-
Buy the book
The Cheater's Guide to Baseball
"Derek Zumsteg should be ashamed to have such comprehensive knowledge of the history of cheating in baseball. Pete Rose gave me two to one odds this book would become a classic." -- Allen Barra, author of The Last Coach: A Life of Paul "Bear" Bryant
Community Projection: Richie Sexson
140 of you guys have chimed in so far, and there was far more agreement in the projections than I expected. Let’s go to the numbers:
Community Projection: .260/.344/.512, 548 AB, 34 2B, 0 3B, 34 HR, 65 BB, 4 HBP, 153 K
Low Projection: .220/.297/.411
High Projection: .281/.378/.606
Dave Projection: .260/.327/.474
Jeff will provide you with all the gritty details in cool graph form, I’m sure, but I find it interesting how close the absolute best case projection is to the overall assessment - basically, the community thinks there’s a really good chance that Richie’s going to have the same year he had last year, and has almost no upside beyond that. There’s a few people that foresee a major collapse, but no one has him performing at an MVP-type level. Basically, Mariner fans think that ‘07 Richie Sexson is essentially going to be a repeat of ‘06 Richie Sexson with some real chance to underperform and not much of a chance to overperform.
The .850 OPS he’s projected to have in Safeco Field is a pretty good offensive performance, but even with that line, he’s still a bad defensive first baseman making $14 million a year with significant downside. They should have dumped him when they had the chance.
Coming tomorrow - the Jose Lopez projection opens its doors.
Comments
113 Responses to “Community Projection: Richie Sexson”
Leave a Reply
You must be logged in to post a comment.
-
USSM uses and recommends
Am I the only one who’s enough of all the Richie bashing around here? He’s one of the most reliable 35-40 homer guys in baseball, he usually gets on base at a better-than-average clip, and he always makes the guy on the mound throw more than two pitches. He’s not free. So what? His contract is perfectly reasonable for a team that has more money than most.
Trading Richie would have solved none of the Mariners problems, and it would have created a gaping hole at first base. And what would you want to fill that hole with? A reliable 35-40 homer guy who usually gets on base at a better than average pace?
Seriously, stop it.
asmario:
They’ve described many different options of what they’d rather spend that $14 million on. If you don’t agree with it, fine. But try giving reasonable arguments to counter theirs then.
“Stop it” doesn’t count.
Richie Sexson is this generations Jay Buhner, He either strikes out or hits a homerun
Seriously, stop it.
If you want uninformed analysis like “he hits 35 home runs every year”, there are plenty of places that will be happy to cater to you. This is not one of them.
Richie Sexson is one of the problems with the Seattle Mariners, and we’ll be pointing this out until he’s not, or until everyone realizes it.
Yay! Responding to myself…
Ok, that being said, maybe I should have had a bit more there to follow my own advice instead of just ending in snide comments… Basically, I don’t think that 35-40 home runs, “better-than average” OBP, and making the pitcher throw more than two pitches (3.9 per pa if I’m figuring that out correctly which is high for the Mariners but not the highest) is worth that much money. Especially if you’re not getting a lot of defense out of it.
If the team had shown the ability to spend money wisely so that they really didn’t need it someplace else I wouldn’t mind a splurge on a 1st basemen or something like that occasionally. But that’s really not the case… They certainly seem to need the money in other places. Not that I think they’d probably spend it any better right now…
[i]Dave Projection: .260/.327/.474[/i]
My feeling is that if his power goes then his average should plummet too. I’m probably wrong.
Aw gee, I f***ed up the tags again. What else is new?
“If you want uninformed analysis like “he hits 35 home runs every year”, there are plenty of places for you to read. This is not one of them.”
Let’s say the average homer is worth about 1.4 runs (and those are definite runs–you can’t hit a home run that results in zero runs). And let’s say the average single is worth about 0.4. To make up for the extra run-production of Richie’s homers, a 15-20 home run hitter would have to hit between .330 and .350, all else equal. That’s why he’s been in the top 10 in baseball in Win Shares twice, and the top 20 four times.
I just think you’re undervaluing Richie’s skills. Yes, he’s being paid a lot, and no, an optimal team doesn’t include Richie Sexson being paid $14 million. But sometimes that’s what you agree to pay for someone who is that likely to hit 35-45 home runs each season while doing everything else at least passably.
It’s not that I don’t follow your arguments, it’s just that I think you’re wrong to let the fact that Richie is slightly overpaid turn you all against his presence on the team with such overwhelming consensus.
Let’s say the average homer is worth about 1.4 runs (and those are definite runs–you can’t hit a home run that results in zero runs). And let’s say the average single is worth about 0.4. To make up for the extra run-production of Richie’s homers, a 15-20 home run hitter would have to hit between .330 and .350, all else equal. That’s why he’s been in the top 10 in baseball in Win Shares twice, and the top 20 four times.
Or, you know, we could use any one of the hundreds of formulas that have been proven to accurately calculate total offensive value. They all come to the same conclusion - he’s an average offensive first baseman.
I just think you’re undervaluing Richie’s skills. Yes, he’s being paid a lot, and no, an optimal team doesn’t include Richie Sexson being paid $14 million. But sometimes that’s what you agree to pay for someone who is that likely to hit 35-45 home runs each season while doing everything else at least passably.
He’s a horrible defender and a horrible baserunner. He doesn’t do anything else passably, much less everything else. He hits home runs, and sucks at absolutely every other part of the game.
it’s just that I think you’re wrong to let the fact that Richie is slightly overpaid turn you all against his presence on the team with such overwhelming consensus.
He’s remarkably overpaid, and the team would be better off without him.
In my home runs vs. batting average example, that’s if the non-Richie 15-20 homer hitter hits the same number of doubles and triples as Richie.
Based on the individual run value of every possible outcome (1b, 2b, 3b, hr, sf, k, etc…), Richie Sexson was worth exactly 1.6 more runs than Kevin Millar last year per 162 games.
Overpaid, yes. But by no means the most overpaid player with “trade value” on the team. Anyone with even a shred of objectivity knows who the one player is who just steals money “hand over fist” when you contrast paycheck with overall production.
Let’s say the average homer is worth about 1.4 runs (and those are definite runs–you can’t hit a home run that results in zero runs). And let’s say the average single is worth about 0.4. To make up for the extra run-production of Richie’s homers, a 15-20 home run hitter would have to hit between .330 and .350, all else equal.
In my home runs vs. batting average example, that’s if the non-Richie 15-20 homer hitter hits the same number of doubles and triples as Richie.
Unfortunately, all else isn’t equal. You have to consider the fact that when someone gets on by any hit, or a walk, or whatever, they aren’t making an out. The average (or OBP) that someone would “need” to have would drop considerably after considering that, depending on how much value is attributed to it.
Other hits (like singles) can also advance runners an extra base, can allow the batter (becoming runner) to steal, etc.
Looking at it exactly as you said it, a player who hit .100 and hit a HR every 10 AB (every hit) would be worth exactly the same as a person who hit .350 with no power at all. Nah. It’s not like there would be a huge difference in production in the end, but it’s clear to me which one is worth more.
That’s why he’s been in the top 10 in baseball in Win Shares twice, and the top 20 four times.
There’s an accurate metric for determining a player’s worth.
nyone with even a shred of objectivity knows who the one player is who just steals money “hand over fist” when you contrast paycheck with overall production.
OK, I’ll bite. Washburn? Bloomquist if you’re talking about ratio of paycheck to production as he’s 2:1, but he’s only overpaid by 500,000.
Sexson is a 3 win player making $14M; I still think he wins.
“Or, you know, we could use any one of the hundreds of formulas that have been proven to accurately calculate total offensive value. They all come to the same conclusion - he’s an average offensive first baseman.”
If by “average” first baseman, you mean that there are ten to fifteen better-hitting first basemen in the league, you could certainly argue that. But even so, the ten or fifteen best hitting first basemen in baseball are among the best hitters in baseball. You wouldn’t call Richie an average hitter, would you? And would you want an average hitter at first base if you were assembling a team? And how much of the budget would you set aside to prevent an average hitter from playing first base for your team?
I’m not arguing that he isn’t overpaid. I’m arguing that it shouldn’t be that big a deal that he’s overpaid. It’s good to have Richie in the lineup at first base, and there’s plenty of non-Richie budget room for a great team. They just don’t use it right.
“Unfortunately, all else isn’t equal. You have to consider the fact that when someone gets on by any hit, or a walk, or whatever, they aren’t making an out.”
Oh yeah, good point.
These projections that have him getting 200 at bats and hitting 10 homers need to be thrown out of the data set or they need to be extrapolated out to 500 at bats.
Predicting a catastrophic injury is just someone trying to be a wise-ass and isn’t in the spirit of what we’re trying to do here.
It’s just like someone predicting that Felix will put u a line of 300ip 320Ks 17BBs 31W 2L 0.41ERA.
Basically, extreme outliers don’t help the data set’s accuracy.
I was wondering what you think about his dramatic pre/post ASB splits.
Pre-All Star (330 AB)
.218/.288/.418
Post-All Star (261 AB)
.322/.399/.613
Do you think he is a consistantly slow starter. If he could avoid the slump he had last year, he’s a very valuable player. It seems like he’s prone to slumps because of his long swing. If he could cut any slump he has down from a 330 AB slump to a 100 AB slump, he is still a good player even though his defense and baserunning skills are pretty bad. I guess I still see potential in him. I also realize he could suck at any moment.
Sexson would be a much better player if he would watch a pitch or two on occasion.
I don’t know if he just got frustrated at umpires calling strikes on him that would be at the knees of a 6 foot guy, but his hacktastic approach drives me nuts.
If by “average” first baseman, you mean that there are ten to fifteen better-hitting first basemen in the league, you could certainly argue that. But even so, the ten or fifteen best hitting first basemen in baseball are among the best hitters in baseball. You wouldn’t call Richie an average hitter, would you? And would you want an average hitter at first base if you were assembling a team? And how much of the budget would you set aside to prevent an average hitter from playing first base for your team?
The whole point of building a team is to be better than your opponents. If Sexson is worse than 10-15 other team’s first baseman, than he’s not a competitive advantage. You don’t pay all-star money to league average players.
His value is relative to his peers. His home runs have no inherent value in and of thsemelves. Players are valuable because they’re better than what your opponent has.
Sexson isn’t.
These projections that have him getting 200 at bats and hitting 10 homers need to be thrown out of the data set or they need to be extrapolated out to 500 at bats.
Do you want me to throw out the projection that has him hitting 51 home runs, too, because that’s even less probable than him getting hurt?
The outliar projections are overwhlemed by the sample we’re taking, and there’s no reason to throw them out.
Or because they’re not too much worse than what your opponent has. You can’t just throw a AAAA replacement player out there and put a checkmark next to first base when your opponents have Pujols, Paul Konerko, Ryan Howard, Todd Helton, etc.
Sexson’s career post-ASB OPS is 127 points better than his pre-ASB OPS. So, yeah, he’s a slow starter.
Dave,
I am not sure if this has been discussed before, but is there any evidence that a last place team has to pay more in free agency for a player than a first place team would. I guess what I am getting at, is that although Richie Sexson is paid grossly more than he is “worth”, perhaps the Mariners would have to pay any player more than they are worth to convince them to come to Seattle during the last couple of seasons.
Although I completely disagree with the sentiment, I’m betting this comment referred to Beltre since he inspires such money-related hatred despite his fairly reasonable contract in today’s market. I would like to think I have a shred (just a shred, mind you) of objectivity.
Player - WARP3 - salary - $ per WARP
Sexson - 7.6 - $14M - $1.84M
Ibanez - 7.7 - $5.5M - $714K
Beltre - 7.8 - $11.5M - $1.47M
By way of comparison:
A-Rod - 7.4 - $26M? - $3.5M
The 200-AB projections aren’t any further out of line than the others. Players get hurt. The chance of Sexson missing 2/3 of the season is pretty low, but it’s nowhere near zero, either.
Just out of curiosity, what is the projection with the lowest and highest OPS (or other measure of your choosing) thrown out? I’ll bet it’s awfully close to the one with them in.
I understand that the sample size minimizes the effects of the outliers. However, there is a difference between projecting him to hit 51 homers in a season of play and PREDICTING that he’ll suffer a 4-month injury.
One is based on performance that can reasonably be projected with a high degree of probability, the other is based on a random isolated event that can’t possibly be predicted.
Really? So people are projecting him to have an 830ab-60+ homer season?
However, there is a difference between projecting him to hit 51 homers in a season of play and PREDICTING that he’ll suffer a 4-month injury.
I don’t see that difference. No one knows for sure what is going to happen next season whether it is a HR blip or an injury. Both things are equally unknowable. (Also, PECOTA’s 90th percentile forecast for Sexson has him at 40 HR. Saying that 51 HR is something that “can reasonably projected with a high degree of probability” is a huge stretch. There’s a big difference between 40 HR and 51 HR.)
I haven’t looked at the spreadsheet, but I’m sure there are plenty of folks who plugged Sexson for 650+ PA, based on his last two seasons, so a couple of 200 PA predictions will help to bring those down to a more reasonable value to reflect the possibility that he could get injured again.
The 200 AB projection probably has him OBPing at about the expected rate anyway, so who cares? It doesn’t hurt the rate stats, and that’s all I care about.
That being said, I don’t think a community projection is the best method for predicting AB totals (or injury probability) at all.
Really? So people are projecting him to have an 830ab-60+ homer season?
There don’t need to be projections like that.
Let’s think about the numbers here. We’ve got 140 people with an average of 548 AB. If there were 10 people who put down 200 at-bats (I don’t have the spreadsheet in front of me) and we threw all of those out, the average goes up to about 575 AB. If you can’t know that Sexson is going to have a catastrophic injury, then you certainly can’t know his at-bat total for the next season to within 25 AB, which is maybe a week’s worth of games. Even throwing out 10 outliers in this sample would give a meaningless change in the AB projection.
Again, I don’t have the exact numbers in front of me, but that’s the sort of thing that Dave is talking about when he’s saying that “The outliar projections are overwhlemed by the sample we’re taking, and there’s no reason to throw them out.” Once you get up over 100 people putting in projections, a couple of nutjobs aren’t going to seriously mess with the quality of the projection.
I think predicting an injury is fine. If you were going to predict Bonds, you’d have to take into account the fact that he could get hurt. Otherwise you’d be overprojecting his production. It all averages out in the end. With projections, getting a good distribution is good I think and tells us much more about what people think will happen.
I am looking for a better year than last year for Big Sexy. I feel that there is a reasonable chance he improves from last year and is an okay fit. I was not a big fan of the signing, not a big fan of him as a player, but he is not a huge problem, and he could easily hit 40 homers and improve his ops numbers this year.
Or because they’re not too much worse than what your opponent has. You can’t just throw a AAAA replacement player out there and put a checkmark next to first base when your opponents have Pujols, Paul Konerko, Ryan Howard, Todd Helton, etc.
Yes you can. The Tigers just went to the World Series with Chris Shelton and Sean Casey.
But that’s not the point anyways. You’re assertion is that Sexson’s contract isn’t a problem because he hits 35 home runs a year. That’s wrong, and you can’t defend that statement, no matter how many times you reiterate.
Al Martin would be worth 14 million a year…you CAN defend that statement, but where do you begin? That is the question
I am not sure if this has been discussed before, but is there any evidence that a last place team has to pay more in free agency for a player than a first place team would. I guess what I am getting at, is that although Richie Sexson is paid grossly more than he is “worth”, perhaps the Mariners would have to pay any player more than they are worth to convince them to come to Seattle during the last couple of seasons.
People love to bring up this point, but it doesn’t matter - the M’s may have to pay more than the Yankees to sign certain free agents, but they don’t have to sign those free agents. There are many, many ways to acquire players, and signing the guy who takes the most money is the worst way of them all.
If Richie Sexson says “I want more money than I’m worth because your team is bad”, giving him more money than he’s worth doesn’t serve to make us any better going forward.
My proof that Richie Sexson’s contract is not a problem:
1. Richie Sexson’s WARP in his two years with the Mariners: 7.4, 5.7
2. Carlos Delgado’s WARP those years: 6.4, 5.4
3. Jim Thome: 1.4, 6.0
4. Todd Helton: 8.1, 5.6.
5. Paul Konerko: 7.2, 6.1
6. Justin Morneau: 1.4, 6.6
7. Adam Dunn: 6.5, 4.5
8. Ryan Howard: 3.7, 8.0
9. Pujols: 10.0, 10.8
This proves that, while Sexson is clearly no Pujols, his performance has been right up there among the other best-hitting first basemen in baseball since signing with the Ms. Partly because,
9. He’s only missed like, 3, games due to injury.
This proves that he’s a healthy boy, despite the fact that one time he ripped his own arm off with his super-powerful swing. His health afterward isn’t so surprising, as the surgery he had prevents a recurrence of the injury in 95% of cases, and most of the people who have it done are weightlifters or trapeze artists.
10. Take Richie’s contract out of the Ms payroll, and they still would have a higher payroll than 16 Major league teams.
This proves that, even without Richie or the money they spend on Richie, the Ms would have more than enough money to assemble a better than average roster.
11. Richie’s contract runs only 2 more years, so the Ms are out from under it when he’s 34.
This proves that the Ms will stop paying Richie big bucks right when his performance is likely to decline in a serious way.
And all together, this proves that Richie’s contract is not a serious problem, no matter how many times I reiterate it.
Also, Sean Casey made $8.5 Million in 2006.
The issue I have with throwing these out is that the question we’re asking is not “what season do you think he’s likely to have” or “what season do you consider the weighted mean of likely forecasts” or even “what would an average 2007 look like if he played it ten times” — the whole idea is for people to make specific projections.
If 20% of people entering projections think he’ll suffer a serious injury and be out for the season, then that’s useful data in this exercise.
Richie Sexson’s WARP…
Everything after this is invald. WARP is useless for multiple reasons, all of which I’ve discussed in depth on the blog. It’s especially useless for Sexson, whom it gives 20 runs (2 wins!) of credit for his defense in 2006. If you ever needed evidence of the absurdity of the FRAR, it’s that right there. There’s not a rational person alive who could even begin to defend a claim that 2006 Richie Sexson was worth two wins with the glove.
You’re just wrong about Sexson. We’ve had this discussion five times, at least, and I’ve gone through every point you can possibly bring up to defend him. He’s an average major league first baseman being paid like an all-star. That’s a problem, whether or not you want to admit it.
If Sexson has these dramatically different pre and post all star splits, can we just bench him till July 1st?
When the M’s had Mike Moore pitching for them, he’d be great 2nd inning on, all the damage was in the 1st. I suggested then that he start games in the 2nd inning.
WARP isn’t the best way to evaluate that list of first baseman because it factors in BP’s bad fielding stats. Also, it’s misleading to put Jim Thome and Ryan Howard on the list when both players played partial seasons. And Adam Dunn is a left fielder. Plus, you leave off guys like Mark Teixeira, Nick Johnson, and Kevin Youkilis who are equal to or better than Sexson for a lot less money.
I’ve never been compelled to post before, but I fail to see where this borderline obsession with first base defense comes from. Isn’t it the least important fielding position? Seems to me that it is usually the position where you hide a guy who can’t field but can rip it. I could be wrong, but I think your anti-Sexson diatribes would find a more welcome reception if you focused on his hitting deficiencies…
10. Take Richie’s contract out of the Ms payroll, and they still would have a higher payroll than 16 Major league teams.
This proves that, even without Richie or the money they spend on Richie, the Ms would have more than enough money to assemble a better than average roster.
Yes, however, it doesn’t prove that Sexson’s contract is good, or “not a problem.” I thought you wanted to show “My proof that Richie Sexson’s contract is not a problem”.
11. Richie’s contract runs only 2 more years, so the Ms are out from under it when he’s 34.
This proves that the Ms will stop paying Richie big bucks right when his performance is likely to decline in a serious way.
He took a pretty big hit last year. If he gets worse, or even stays at that level, is that not a decline? The contract wasn’t up after 2005.
“borderline obsession with first base defense”?
It’s not as important as shortstop, but needs to be considered. It doesn’t go away because it’s comparatively less important.
Borderline obsession?
I’ve never been compelled to post before, but I fail to see where this borderline obsession with first base defense comes from.
I don’t think it’s an obsession. I’d say it’s mainly the opposite. Saying that Sexson is this good, including 2 wins from defense alone, would be overstating it even if he were good. It’s simply not that easy for any first baseman to be worth that much defensively.
So in essense, most people are probably agreeing with you, at least to the point that it isn’t that important how good a first baseman is defensively, unless he’s really on either extreme.
That’s what I see being posted, anyway. I don’t see anyone saying that he’s “worth” a loss of two wins defensively.
An absolutely obvious statement but one many, many fans fail to grasp.
Case in point-Adam Dunn. He hit 40hrs last season. Yet, he was dramatically overpaid even at the paultry $7M he collected last season. He only managed a VORP of 23 (2 win bat) which was almost entirely wiped out by his atrocious defense (he was somewhere between a -10 to -25 defender according to a survey of the gold standard defensive metrics). At best he was a 1 win guy at the atrocious rate of $7M/win.
The going rate for a win these days is something like $3M to $3.5M.
There was a great thread discussing Sexson’s worth posted during the holiday season….it completely lays out the argument for why Sexson is not a good value. Basically he is a 2 win player being paid $14M. That’s an obscene $7M/win.
Still not convinced that $7M/win places a team at a competitive disadvantage? Consider the Reds again. They platooned Hatteberg and Aurilia at first for a cost of $2.25M. The dynamic duo totaled a VORP of 44 (roughly a 4 win bat). Their defense was somewhere around neutral. So the Reds solved their first base dilemma for about $500K/win. That is a bargain. In other words, the Reds paid $12M less for a net of 2 additional wins over what Sexson is projected to do in ‘07.
Just because its one of the easiest positions, doesn’t mean that playing it poorly is a benign thing relative to a player’s total contribution to wins… See the Dunn analysis in #46. The same argument could be made about leftfield being easier to play defensively. Yet poor defense completely nullified Dunn’s value in ‘06.
“WARP is useless for multiple reasons, all of which I’ve discussed in depth on the blog.”
Oh, I see. You want people to pour over your blog archives, not comment on your current blog posts. I guess I misunderstood the premise of your medium.
It’s pore, by the way, not pour.
I guess it’s too much to expect you’d look that up either.
Oh, I see. You want people to pour over your blog archives, not comment on your current blog posts. I guess I misunderstood the premise of your medium.
I linked to one of the comments where I explained why WARP is useless. That search took me about 8 seconds.
If you missed all the other Sexson conversations, I’m giving you the cliff notes - Sexson’s an average first baseman being paid like an all-star. If you want to see the evidence of this assertion, I’ve already written it, multiple times. I’m not going to rewrite the entire argument for you when googling for it is so, so easy.
I’m pretty sure he meant he thinks you want people to pour the last drops from those old beer cans… um, oh yeah, M’s don’t want us to remember that.
Dave/Derek: Is it fair to say that the issue you have with WARP is that replacement level for defense is roughly league average (actually probably a little slightly above league average) but WARP sets replacement level much lower thus over inflating wins? To me, the way WARP is calculated sounds like it was partly constructed to avoid the Jamesian argument that you can’t sum things normalized to different baselines (i.e. it’ll give you a cardiac infarction if you attempt to sum VORP and a defensive metric that uses league average as a baseline because well thats just illogical).
This really isn’t a problem though since league average IS replacement level for defense…are we on the same page?
I guess I see winshares as hopelessly flawed because it can’t handle defense and WARP hopelessly flawed because it decided to fumble on defense so it could date winshare’s cheerleaders.
WARP has many, many problems. The main one is that replacement level is set ridiculously low, since offense and defense are calculated separately, so that Wins Above Replacement Level essentially means Wins Above A-Ball Bench Player. Since no one cares about Wins Above A-Ball Bench Player, the number ends up being meaningless.
It also includes the DT fielding translations, which are horrible.
Win Shares isn’t any better. Both systems are basically useless.
This has nothing to do with our overpaid, average first baseman but, did anybody read Caple’s aticle about the Aaron comeback? Oh man, there are some classics salvos fired right at the M’s. Check it out. http://sports.espn.go.com/espn/page2/story?page=caple/070208&lpos=spotlight&lid=tab3pos2
But Al Martin guys….u cant compare any overpaid player without talking about the most underpaid player in Al Martin…the guy was amazing…amazingly ugly, but amazing all the same
I love this site for this very reason. There is nowhere else that one can find this level of analysis or projection. I loved last years cookie bowl. Will there be another local get-together before opening day? I’d be happy to help.
Why exacty are DT translations bunk?
Dave: If the bottom line is A-level instead of, say, AAA-level, then the standard deviation will be vaster and hence the 50% level be harder to pinpoint and the 10% levels soar and plunge more acutely. Not meaningless though. Just harder to say anything remotely intelligible about.
That’s not what I said. I said that performance is something that can be projected with a reasonable amount of certainty. I did not say that projecting him to hit 51 homers is something that you can do with a high degree of certainty. I fully understand that 51 homers is a highly unlikely event.
I guess my problem is that I believe that there is a line between projecting and predicting. One uses data, the other uses a wild ass guess. Why not predict that some unforeseen event is going to cancel half of the baseball season and project all the players to have half seasons? It just seems silly to me. Especially when you’re talking about a player that has suffered one major injury in an otherwise really healthy career.
Why exacty are DT translations bunk?
Evaluating Defense” discusses this in depth.
Because of the misapplication of replacement level, WARP overvalues all players, but for the most part it does so equally. In theory that’s correctable by looking at FRAA instead of FRAR. Except that BP’s FRAA just isn’t very good. And that’s really unfortunate because BPs stats and PECOTA are certainly the fastest way to look up how many wins a player was worth.
On outliers, 51 HR is ~3 std deviations above the mean while 200 AB is 5 std deviations below the mean. Of course we wouldn’t really expect a bell shaped curve — you can only improve so much but it’s always possible to fall off to zero. And in 150 predictions, one really doesn’t matter. The only ones worth tossing are if someone is screwing around say 600/120/5/0/45/60/210 or 600/200/40/1/65/90/110.
You bet wrong. It really shouldn’t be that tough to figure out.
Actually, no it doesn’t which is why I asked. Basically DT is usually dismissed because of a lack of transparency though Dave indicates there are many things wrong with it. I’ve always been curious about those mysterious many things. What specifically?
But the misapplication of replacement level only applies to the defensive side, right? How BP handles the offensive side isn’t being criticized as a flaw of WARP here is it?
In MLB’s ineffecient free agent market, HRs are very valuable in and of themselves (to the guy getting the contract, of course). I think that the majority of M’s fans would probably side with asmario on this one, and I think that most front offices still think that HRs put butts in the seats. Moreover, do you know how many female Mariners fans have Sexson jerseys? You should have seen my wife when I told her that Sexson was actually hurting the team.
SHUT UP, I’M NOT BITTER!!
If the majority of front offices over-value HRs, then a SMART front office will place value on other things that those other front offices under-value, thereby getting a competitive advantage. That was the whole point of Moneyball, not the “OBP is king” thing. Of course, we have scant evidence of any intelligence in the Mariner front office these days, and thus they keep Sexson around.
In light of the heated discussion on Sexson’s merit, I had a thought… how would you think the Mariners would fare with, say, jettisoning Sexson and using Ben Broussard as the everyday 1B, maybe even platooning him with someone like Bryan Lahair?
Help us out, here, Ralph. Who do you think it is?
Gomez - It would have abeen a better idea 4 months ago when there was still a chance to use Richie’s salary on someone else (like maybe some pitching).
Ralph likely thinks that it is Ichiro that is vastly overpaid. Now, Dave is unlikely to like it because it uses WSAB, but the Hardball Times Marginal Wins Calculator had Ichiro as UNDERPAID based on performance last year, but a couple million dollars.
As for Sexson, just to add another defensive measure to the mix, Defense Runs (controversial system yes) has him at basically average, less than +1 run and right in the middle for Firstbasemen used in any significant frequency last year.
Dave and Derek, I’m not trying to be a link whore, but here it is
ralph loves WFB
Assuming you are talking about Ichiro…
Ichiro is more than 50 runs over a replacement centrefielder offensively, and probably at least another 10 on defense. That’s 6 wins for around $13 million.
THT reckons that free agents give a return of about 1 win for every $4.4M. He still underpaid now that we’re tried to have an objective look at the situation?
If he’s 6 wins better, yes. 6*4.4M= 26.4M$ on a contract of 13M$
just read that we signed Jamie Burke…any chance he makes the club as the backup backstop so Rivera can move down and actually improve with consistent playing time?
It’s especially useless for Sexson, whom it gives 20 runs (2 wins!) of credit for his defense in 2006.
Richie is stealing Yuniesky’s FRAR!
Seriously, though, I don’t know how Clay Davenport can tell the guys at BP that his defensive stats are worth anything when there are huge, huge outliers like that everywhere.
Okay, I’ve pored (thank you) over some of your previous posts on the Sexson subject, and I still disagree that his contract is a problem. And if it is a problem, I think the best solution is to ride it out, at least until a trading deadline when we might get a valuable prospect for him. Even then, I’d be shocked if we could trade him to anyone without continuing to pay him at least $4 - $6 million per season.
You’ve probably gone into this before (I did an 8-second Google search and couldn’t find anything), but do you believe it’s reasonable to pay every player based on his expected performance without paying extra for a higher lower-bound or a smaller standard deviation? Just looking at a player’s expected performance, no matter what metric you prefer to use, against an average performance strikes me as a mistake when most players are more likely to suck horribly, any given season, than a Richie Sexson is.
It’s reasonable to believe that every modicum of probability that a player will not suck has value, and we can be more certain that Richie will not suck than we could be with the vast majority of possible replacements. Does this paying-for-probability cause Richie to be paid more than his expected performance (viewed as one number) can justify? Yes. Does that make his contract a problem with the certainty that you insist? Nope.
Guys with semi-decent sticks at 1B just aren’t that hard to find for a lot less than 14 million.
Case in point:
Sexson, 2005: 264/.338/.504
Player A, career: .266/.328/.465
Cost to acquire Player A: Shin-Soo Choo and $3.55 million.
Is Richie Sexson 10-11 million more valuable than Broussard? I’d have to say “No”.
Hello everyone, I haven’t posted for a while, but I’ve been meaning to ask you guys this and I apologize if this is discussed elsewhere, but has anyone done any sorts of calculations with regards to Richie’s “clutchiness”? I know that those sorts of analyses are often contentious, but I was speaking with a friend the other day who pulled out some stat like “With runners in scoring position, Richie’s average goes up .02 points” or something. Anyway, just curious to know what y’all thought. I haven’t done the analysis myself, but it does seem like Richie will often come through in the clutch.
Godliness > cleanliness > truthiness > clutchiness.
asmario - while few would agree with your arguments, I find parts of it valid, and admire your presence to follow through, get more details, and contribute to the spirit of the debate while staking out a minority position.
Richie Sexson’s stats is worth almost +2 wins as a hitter above average, considering he bats in Safeco half the time. USSM readers see him as a slightly-below-average fielding 1B, and I give a positional adjustment of -1.0 wins for 1B. So, he’s worth around 0.5 wins above average, maybe a bit more. That’s +2.5 WAR for 2007, +2.0 in 2008, and +1.5 in 2009. The free agent dollars per win in 2007 is 4.0, in 2008 it’s 4.4, and 2009 likely 4.8. Multiply all that, and you get a total free agent worth of 26 million$.
Does he have about 46 million$ coming to him? Under that scenario, if he was +4.0, +3.5, +3.0, then he’d be worth 46 million$.
That is, we should properly value him as +2.5 wins above replacement. The Mariners on the other hand are valuing him at +4.0 wins above replacement. That is an enormous gap.
In order to justify that, Sexson would have to hit as well as he does, while play an average THIRD base.
I wasn’t clear: +2 for hitter, -0.5 for fielding, -1.0 for positional adjustment. That’s +0.5 wins above average. Average is 2 wins above replacement.
Why exacty are DT translations bunk?
Because they simply don’t contain enough real information to judge what they’re attempting to judge. Non-play-by-play metrics have a lot of challenges to overcome, and Clay’s metric doesn’t do very well.
But the misapplication of replacement level only applies to the defensive side, right? How BP handles the offensive side isn’t being criticized as a flaw of WARP here is it?
Offensively, BRAR and BRAA are fine - they’re basically just another version of VORP.
You’ve probably gone into this before (I did an 8-second Google search and couldn’t find anything), but do you believe it’s reasonable to pay every player based on his expected performance without paying extra for a higher lower-bound or a smaller standard deviation? Just looking at a player’s expected performance, no matter what metric you prefer to use, against an average performance strikes me as a mistake when most players are more likely to suck horribly, any given season, than a Richie Sexson is.
Performance certainy has value. There’s no reason to think that Richie Sexson has any more performance certainty than other major league average first baseman, however. You’re ascribing him a skill he does not possess.
Does this paying-for-probability cause Richie to be paid more than his expected performance (viewed as one number) can justify? Yes. Does that make his contract a problem with the certainty that you insist? Nope.
So now it’s not Richie’s 35 homers that are valuable, like you initially stated, but instead it’s your belief that he won’t totally suck, and your belief that he’s unique in that regard, that makes him worth $14 million a year?
That’s crazy. Baseball is littered with guys who can perform at 95% of Sexson’s abilities for 10% of the cost. Even players who perform better than Sexson are being rewarded with contracts at a fraction of the cost in a market where sanity went out the window.
Among major league regulars, there are few players who help their teams less than Richie Sexson pulling in $14 million a year.
In response to post 81:
Yes, it is still his 35-45 homers that are valuable, and it’s the unusually high probability of those 35-45 homers (along with the probability he gets on base at a .370 clip) that make paying Richie well an uncrazy thing to do–or at least a less-crazy thing to do than you say.
Yes, baseball is, each year, littered with first basemen who perform at a 95% Richie-Sexson level for much less money, but which of those players will do so next year, and which ones will perform at a 60% Richie-Sexson level? Each one is a gamble. Richie is much less of a gamble than most first basemen.
Not to derail this endlessly enlightening Richie Sexson discussion, but at our current rate of projections we will finish the 14 starters after seven oh, so meaningful ST games have already been played. (I’m assuming 3 projections per week and weekends off).
How are the authors going to counteract the influence of Horacio Ramirez’s breakout brilliance as he pitches a 4 inning no-hitter in his first half-game and we project him the next day?
Are we projecting WFB this year, and if so how can we be objective after he grits out two in the park home runs AND steals three bases against our hated rivals, the Padres?
I’m worried that future generations will be able to see our terribly tainted little experiment and we will only do harm to the reputation of the statistical community.
Let’s pick up the pace!
Also, I’m not ascribing him a skill he doesn’t possess. I’m saying that a guy with a major-league performance record like Richie Sexson’s, proven statistically significant over the course of many seasons, is less likely to be a scrub next year than a guy like David Dellucci, who has spent statistically significant portions of his career being a scrub.
Richie is much less of a gamble than most first basemen.
No, he’s not. PECOTA gives him a 34% collapse rate and an 11% attrition rate for 2007, among the highest of any first baseman in baseball. The only regular first baseman with higher collapse rates are Nick Johnson, Todd Helton, and Derrek Lee.
You’ve given no reason for us to believe that Sexson has any more performance certainty than any other first baseman, and empirical data refutes your claim. There’s really no reason for us to believe it’s true.
Also, I’m not ascribing him a skill he doesn’t possess. I’m saying that a guy with a major-league performance record like Richie Sexson’s, proven statistically significant over the course of many seasons, is less likely to be a scrub next year than a guy like David Dellucci, who has spent statistically significant portions of his career being a scrub.
Okay, then, you’re just wrong.
[sigh] OK, I was trying to derail it before it derailed itself. Too late.
Post 85 & 86: My reason for believing Richie has more performance certainty is the law of large numbers. I don’t buy PECOTA’s collapse rates, and in the real world it’s not possible to be wrong about something that hasn’t happened yet.
Alright, I’m done now. Actually doing my job is starting to sound kind of nice.
My reason for believing Richie has more performance certainty is the law of large numbers. I don’t buy PECOTA’s collapse rates, and in the real world it’s not possible to be wrong about something that hasn’t happened yet.
This offseason must have been a thing of dreams for you, then. Ignoring empirical data and believing in veteran value due to performance certainty basically defines the Mariners player acquisition strategies. It’s this kind of thinking that keeps the Mariners in the market for name players with past successes, and it’s this kind of thinking that makes us the worst team the division.
Not at all. Jose Vidro’s deep performance record is what makes it clear he’s not a good hitter. Performance certainty only makes a player valuable if his performance is good. I believe Richie’s performance has been good, you don’t. There we are.
Guiseppe - It was a good effort though.
What? That doesn’t make ANY sense.
Yes, it is still his 35-45 homers that are valuable, and it’s the unusually high probability of those 35-45 homers (along with the probability he gets on base at a .370 clip) that make paying Richie well an uncrazy thing to do–or at least a less-crazy thing to do than you say.
Number of times Sexson has hit 35-45 HRs since 1999 (8 seasons): 3
Number of times Richie has OBPed .370 since 1999 (8 seasons): 1
The only way the law of large numbers is relevant — if i understand that law correctly — is if you believe Richie Sexson will have the opportunity to play his 2007 season a large number of times.
If we’re really going to do the Groundhog Day thing, can we please do it with the 2001 Mariners and not the 2007 Mariners?
I think we’ve been doing the Groundhog Day thing the last 3 years.
Player performance can decline quickly. Were you not a Mariner fan when guys like Bret Boone & John Olerud were DFA’d? They certainly had no prior history of being scrub players. Using your theory, keeping these guys was great strategy because obviously their past performance made them less risky.
That’s not to say I think Sexson’s performance is going to fall off a cliff this year, but it certainly could. By overpaying Sexson relative to his performance, the Mariners are assuming significant extra risk. Better to pay top dollar to players who carry less risk of performance decline.
“Number of times Sexson has hit 35-45 HRs since 1999 (8 seasons): 3″
Number of times he’s hit 34-45 since 2001 (6 seasons, 1 missed to injury): 4.
“Number of times Richie has OBPed .370 since 1999 (8 seasons): 1″
Richie’s last 4 park-adjusted OBPs (complete seasons only): .364, .378, .391, .352
One thing: the suggestion that the M’s could do something better than Richie Sexson with the $14 mil presupposes that they wouldn’t just blow it on ten Willie Bloomquists (or one Jarrod Washburn). Looked at that way, having some of the money locked up for a guy who’s probably not going to out-and-out suck is some kind of small comfort, isn’t it?
So now we’re using park-adjusted numbers…and adjusting the home run target down from 35 to 34…way to be a moving target.
If we are gonna talk about 14 million as a waste on Richie (being overpaid) how do you talk about the combined salaries of Washburn, Batista, Weaver, and Vidro and tell me all of that money could not be better spent?
actually this thread has ALOT going for it….
Thanks.
Since we’re still with Mr. Hargrove,lets try to reign back the absolute certainty that some of these refutations come dressed with. You have been wrong before and recently.
Why the law of large numbers is relevant to this discussion:
Each time an unknown or average baseball player comes to the plate, there is a certain probability that he’ll make an out, or hit a home run, etc.
Each baseball player has a certain amount of talent. The meaning of talent, at least in sports, is the amount by which a player’s performance probabilities deviate from the norm. How do we know what a player’s performance probabilities are? They reveal themselves slowly, over time, as the data points accumulate. The more times they hit, the more their observed performance rates converge with their true talent, which previously existed only in a platonic-form kind of state, somewhere only a scout could see it.
There’s a lot of grumbling over Sexson’s annual salary. Isn’t that only a problem if the $14 million would be spent wisely elsewhere?
As guys who constantly criticize the front office’s policies and practices for talent acquisition, what makes you think that getting rid of him would leave the Mariners in better short-term [i]or[/i] long-term shape?
Why the law of large numbers is relevant to this discussion:
Each time an unknown or average baseball player comes to the plate, there is a certain probability that he’ll make an out, or hit a home run, etc.
Each baseball player has a certain amount of talent. The meaning of talent, at least in sports, is the amount by which a player’s performance probabilities deviate from the norm. How do we know what a player’s performance probabilities are? They reveal themselves slowly, over time, as the data points accumulate. The more times they hit, the more their observed performance rates converge with their true talent, which previously existed only in a platonic-form kind of state, somewhere only a scout could see it.
WHich is, of course, an overly simplistic model which ignores several factors. One of the biggest is aging and physical deterioriation. Which is what Dave is pointing out.
Or do you REALLY think that you can project this into age 45, 46 and 47 with absolutely no changes?
David Dellucci has 2,294 major league at bats. I think that’s enough to judge his talent level. That has nothing at all to do with the law of large numbers.
#101 For the sake of objectivity, pretty much everything I have read on this blog has indicated that the Mariners front office is basically incompetent and at least in regards to Batista, Washburn and Vidro, they are way overpaid.
Yes, so a theoretical GM With A Working Clue Phone could put the Sexson money to better use, but the actual GM We’re Stuck With probably couldn’t. So there you go. Enjoy your 34 homers and try not to think about it too much after you’ve had a few beers. That’s approximately 10 minutes of cumulative happiness in what is otherwise going to be a long season.
mmm, beer and baseball..i cant wait!
hmmmm….using a structured system of projecting performance and relating it to salary to determine a player’s value verses worshipping at the alter of the homer,skirting the lack of a true system by discussing *the law of large numbers*, and seemingly ignoring payroll altogether….
While Poisson and Bernoulli are two of my idols (really, i so want to be them)and chicks do dig the longball, i’ll error on the side of the structured system maybe just this once…
106. gwandung
Bring something to the discussion or just read.
yes…you really weren’t just thinking that….your fingers were moving too…
I don’t think anyone is arguing the Pecota is destiny.
However, this thread does have some compelling arguments supporting the conclusion that Sexson is a lousy value given his contract and projected performance.
You be the judge concerning how damaging it is to have roughly 15% of payroll dedicated to paying one player twice the going rate for performance.