Playoff musings
I was looking at the standings after today’s game, and there’s something interesting afoot. The division races are pretty much locked: either Cleveland or Detroit will win the AL Central unless something dramatic happens: they’re both way better teams than Minnesota. The M’s could still catch the Angels, but skip that for a second.
In the wild card, the straight listings run:
Detroit 61-49, .555
Seattle 60-49, .550 -0.5
New York, 61-50, .550, -0.5
Minnesota 57-53, .518, -4
Detroit should really be “AL Central runner-up, .560 or so”. Whoever wants to win the wild card will likely have to put up a record that would win the AL Central.
Anyway, here’s the same race, but with expected win-losses based on runs scored and allowed (from mlb.com):
New York, 68-43
Detroit 61-49
Minnesota 57-53
Seattle 54-55
Obviously teams don’t get credit in the standings for how many games they should have won or lost: the only thing that matters from here on out is how well they play for the remaining games. But it’s hard to look at the Yankees, who started horribly and have played extremely well for months now, and wonder if there’s another team besides Detroit in that bunch that can keep up. The Yankees offense is putting up almost six runs a game (seriously — 660 runs, 111 games, 5.95 runs a game)(M’s are at 4.83). For all the abuse the pitching’s taken, it’s in the middle of the AL pack, and when you score six runs a game… yeah.
There’s the problem. With 50 games left to play, if the Yankees keep this up, they’ll win another thirty games. The Tigers/Indians can put that together, too, so the wild card question becomes “can the M’s keep up that pace?”
That’s tough. Which is why it may actually be easier for them to take the division. Really. Here’s their schedule against their potential competition for playoff spots:
August 27th: 3 game series against the Angels
August 30th: 1 game makeup in Cleveland
September 3rd: 3 games in New York
September 7th: 3 games in Detroit
September 20th: 4 games in LA
September 25th: 4 games against Cleveland
The Angels might be as good as Detroit/Cleveland, but there’s only one of them. The M’s get seven games against the Angels, and if they go in with a 3.5 game gap in the standings… that’s where the opportunity is.
This is why you can look at the postseason odds right now and see the M’s playoff chances through the division title are about four times better than the wild card. Many things have to break right for the M’s to see their wild card competition fall away, but the Angels… well, it pains me to say this after this week’s games, but I hope Boston hands them an even more embarrassing beating than we had to endure.
With 53 games left, how many Wins are needed? Finishing 35-18 gives the M’s a record of 95-67. The team could still win the division without that many wins. It could be a very interesting last week of September.
What about taking last years numbers into account?
M’s had their worst record last August and were quite decent in September
Since both 06 and 07 June are the hottest month, maybe we really have a shot next month?
There is no way in hell the Mariners go 35-18 the rest of the way. We’re barely wavering around 10-13 games over .500 after 109 games. What makes anybody think that we’ll be able to suddenly surge forward to 17 games over .500?
Then again, I don’t think the Yankees fulfill their Pythag either. By all rights, accounts, and purposes, they should have been doing that all along. They haven’t been. I’m not sold on Pettitte’s improvements. I’m pretty sure everybody’s figured out Roger is not the force they thought he was. Hughes may be talented, but has to do it, and Karstens sucks.
All I’m saying is that once the entirety of the Yankee lineup stops being hot like all teams go through, we’ll see just how much they keep it up.
The Yankees September schedule is pathetically easy. They finish the last two weeks playing either TOR/BAL or TB. Ugh.
http://newyork.yankees.mlb.com/schedule/index.jsp?c_id=nyy&m=9&y=2007
It’s August. The Ms are in the thick of it. Why then yesterday’s lineup?
Is it the talent that will do the Ms in or will it be the management of clubhouse personalities?
I would like to point out that the M’s Pythag is skewed, as they all are. The M’s, for example, are 13-16 when either team wins by 5 runs or more, 8-11 at 7 runs and a whopping 0-5 when the differential is 9 runs or more! 0-5! Thats 45 runs! That number will skew any projection.
Just something I thought I would throw out there
I think the best we can hope for right now is an Angels collapse with the way the Mariners are being managed right now. Bavasi and McLaren are far too worried about hurting people’s feelings instead of fielding the best team they possibly can.
And the Yankees pythag is skewed in the other direction. They’ve been winning in numerous blowouts, but not winning as many close games.
Pythag projections are OK as far as they go, but +/- five games is well within the bounds of normal, and systemic factors can add to that.
And that being said, the M’s have been a .500 team under McLaren, and since the AS break. That’s pretty miraculous with half the lineup hitting for a sub-.600 OPS and various pitching meltdowns. It’s going to take some serious buckling down and actual good performance to finish with better than about 88 wins.
Did anyone actually think the M’s would be where they are now at the beginning of the season? Seriously. All M’s fans should be pleased with the progress the M’s have made this year and within the system. Fighting for a playoff spot in August in the AL is something the M’s and their fans should be proud of.
Looking forward, Anaheim faces Bostons 3-4-5 starters and they miss Santana in the series against the Twins.
We miss Bedard vs the O’s, Buerhle against the White Sox, but we hit Sanatana against the Twins.
Is there a statistic that measures the quality of pitching a team faces? It seems like eyeballing the schedule we hit a lot more #1s than the Angels. Man, I wish the Angels could know our pain of facing Dice-K and Beckett on back-to-back days.
#9:
Did anyone actually think the M’s would be where they are now at the beginning of the season? Seriously. All M’s fans should be pleased with the progress the M’s have made this year and within the system. Fighting for a playoff spot in August in the AL is something the M’s and their fans should be proud of.
meh. The performance of all teams fluctuates from season to season. Some years a few more players have good years than bad years, one or two players from whom little was expected become solid players, and/or a team gets more than it’s share of breaks. Then the team posts a good year. Other years the obverse happens and the team plummets. The rest of the time they play somewhere near their proper expectation.
You can’t assess the direction of an organization just based on one year’s performance. The reality is that the Mariners have consistently been a large payroll team since they moved into Safeco. They have had ample resources to put a competitive product on the field.
Despite those resources, the overall record of this regime is mediocre at best. The baseball operations staff is not as bad as the team’s performance has been the last several seasons – a good measure of team improvement is simple regression to the mean, the mean being the Mariners proper status somewhere around the middle of the pack in baseball. This year the team is having an up year because some pieces have meshed unexpectedly. But the context behind that is that the Mariners continue to make the same types of decisions that drag the team into the middle of the pack and negate the teams revenue advantages.
*******
If an organization has skilled management and adequate resources, that organization will produce consistently strong results. Conversely, mediocre management with ample resources almost invariably yields mediocre results.
The Mariners have ample resources; their record with those resources points strongly to mediocre management. It will take more than one year in which they are scrapping to make a playoff spot to indicate any kind of a trend or to suggest the USSM denizens have been totally offbase in their critcisms and analyses of the Mariners front office.
I would like to point out that the M’s Pythag is skewed, as they all are. The M’s, for example, are 13-16 when either team wins by 5 runs or more, 8-11 at 7 runs and a whopping 0-5 when the differential is 9 runs or more! 0-5! Thats 45 runs! That number will skew any projection.
Just something I thought I would throw out there.
You’re saying it’s a positive that the M’s are blown out? Generally speaking, good teams do the blowing out.
And yeah, the Pythag is skewed, which is an indicator that this team is lucky as hell to be where they are. Realistically, this is an 80-85 win team that’s gotten good breaks in their favor (the team’s stayed pretty healthy and so on) that have pushed them closer to 90, but I’m unconvinced it’ll stay that way.
Oh, and the problem I see in matching up with the Angels is they are very, VERY RHP.
The fact that we’re having this conversation in August is wonderful, and they don’t hand out trophies based on Pythag results, but what happens on the field, so I’ll happily take all the luck I can get, all the way to the playoffs and World Series if need be… but I don’t see this team having the hallmarks of a 90-95 win team in and of itself. The management isn’t putting a good defensive team on the field a lot of the time, the team is bound and determined to make sure Raul, Turbo and Richie are completely useless before they make lineup changes, the starting pitching behind Felix is a bunch of 4th starters who are vulnerable to good hitting teams, and so on.
What worries me is if the M’s go 89-73 with a Pythag representative of a 82 win team, and finish a couple of games out, the front office will go “Yay! Bavasi’s proven his worth as a GM!”… when in reality all that will have happened is that we will have gotten a hot roll at the MLB craps table one season, and the offseason decisions will be skewed into thinking this is really a contender instead of a decent but flawed team.
This seems way too close to the 2002-2003 seasons, where we were just that genuine fifth place team. Second in the AL west to the A’s, Second in the wild card to he Red Sox. Neither of those teams slowed down more than the M’s did.
12 – Pythag is a decidedly poor measure of luck as far as I’m concerned. There are so many things that can skew it – a terrible starting pitcher given a prolonged run in the rotation, a streaky offense, etc – that it’s not a precise measure of anything. I’m not picking on what you said necessarily, because lots of people try and use run differential in this way, I’m just saying we need to take it with a pinch of salt. It’s not a sophisticated statistical measure, more an article of faith.
The anti-pythag crowd has taken a tiny shred of truth and turned into a ridiculous claim. Pythag is an excellent predictor of how a team will play in the future, and the “wahh! blowouts skew the point!” argument is ridiculously overblown.
The Yankees are a vastly superior team to the Mariners, and their pythag is picking up on that fact.
Dave you can use all the stats and figures you like but me I wait til the clubs play and see if the human factor kicks in if it does then you see the most interesting upsets happen.
All the numbers and the stats etc… say this team should sweep this one but the opposite happens the sweepor at least the series win goes to the predicted loser.
Yes the Yankees are on paper and when the y play well a superior team not just tothe M’s be to any club in MLB.
However I still consider them noting but a bunch a paid hired bats and gloves who earn their money by winning.
Even with all that money Steinbrunner has poured out he still fails,still loses games,and the human factor stills bites him in his rear and wallet.
As to the playoff musing the M’s have aexcellent shot at the wildcard berth in the Al with outside shot at the division.
However the pivotal series are the following in both scenrios:
1)The remaining 7 games with the Angels Home and home
2)Sept away series with the Tigers
3)finally the Aug 30th game and the Sept 25th home series with Cleveland
Should the M’s manage at least win on the Aug 30th game and at least have 3 out 4 series wins in those contests the M’s will have their playoff shot.
17 – I can’t wait to see this “human factor” in action. Are you suggesting the Mariners have superior humans to the Yankees?
17 – Yes, you can see the most interesting things happen. But you have to get to the playoffs first for that to happen. And almost always the teams with the best pythag make the playoffs. Because pythag is an excellent indicator of a team’s success.
I believe the last team with a negative pythag to make the playoffs were the San Francisco Giants in 1997. They scored 784 runs, allowed 793, yet somehow went 90-72. And the myth of Dusty Baker was born.
The Mariners right now are -3 in their pythag. For comparison sake, the Mariners in 1998 had a +4 pythag. That team went 76-85.
Also, playoff wise, I have to assume the Yankees will have control of the wild card race by the end of the month.
“However I still consider them noting but a bunch a paid hired bats and gloves who earn their money by winning.”
Not sure that NY being a high payroll team does your point any favors; the M’s have fielded a lot of VERY expensive mediocre teams and thrown a ton of cheddar at so-so free agents — The Yankees have high payrolls, but they aren’t throwing it at the Scott Speizios and the Ho-Rams of the league.
I don’t know what you’re getting at there with the human factor comments…it’s not as if the Ms “want it” any more than any of the other contending teams do. This team has a shot at the post season, but I think they’re going to need some help from other clubs to get there.
16 – It’s a point of logic that a team which scores more runs than it gives up is likely to have a better record than a team with the opposite run differential, but using it to accurately predict what a team’s win-loss record will be is an exercise in confusion. It undoubtedly tells you something, but using what is essentially an unaltered set of raw data to precisely predict wins isn’t very scientific.
Moreover it’s a decidedly arbitrary notion that the only influence of any real significance in this scenario is “luck” and not other factors. I have yet to see any conclusive evidence that other factors aren’t the largest influence on the deviation of actual win-loss records from those predicted. If we’re going to make a case that this is accurately measuring luck, then we have to have some real evidence for why that is. It’s not a precise statistic as it stands, we’re essentially saying that blowouts, subtle differences between the way lineups are constructed, pitching outliers, clutch performance and any other influence you care to think of, all “even themselves out” and the only forces of any significance are quirks of circumstance.
That’s a completely arbitrary notion that’s become an article of faith – just look at how equivalent statistics are treated in other sports; in soccer for instance the idea that you could accurately predict a team’s place in the standings on the basis of goals scored/lost would be roundly ridiculed because there’s a general acceptance that some teams are constructed in a way which lets them win big in comparison to other, technically equal, teams.
but using what is essentially an unaltered set of raw data to precisely predict wins isn’t very scientific.
No one is doing that.
Moreover it’s a decidedly arbitrary notion that the only influence of any real significance in this scenario is “luck†and not other factors.
The overriding factor in pythag variance is the distribution of runs. All the research done on the subject shows that run distribution is about 95% random and about 5% skill.
Read this if you haven’t already. You can go through all the machinations necessary to account for the diminishing returns involved in the 5% of run distribution that is skill to adjust pythag for the nuances that it doesn’t catch, and the difference is still tiny.
If you want to argue that distribution of runs is predictable, you’re going to be arguing against a mountain of evidence. If you don’t want to argue that distribution of runs is a skill, then you essentially need to concede that deviation from pythag is mostly luck.
However the pivotal series are the following in both scenrios:
Um, Yankees series?
The big problem I have with winning the division is that we haven’t picked up any ground in a month (we’ve gone 13-13, the Angels have gone 12-12)… which means less time to knock out their lead. The Angels aren’t wasting time with Shea Hillenbrand as their DH any more, and we’re busy shooting ourselves in the foot playing bad defensive and offensive veterans.
With the Pythag record, its just an indication of what the probability is. There is really no other way we can just look at the rest of the Mariners schedule and just say, “Hey, we win this game, lose that win, win this series….” The record helps up with predictions. Just the fact that there are so many games left will take care of some of the variance (Luck, poker term). We all understand that Pythag records don’t mean anything, thats why we play the games. Things happen. However, the Pythag record has been good indicator.
Keep in mind your Pythag changes as the season wears on, since it’s based on the sabermetrically complex factors known as ‘runs scored’ and ‘runs allowed.’
That’s a completely arbitrary notion that’s become an article of faith – just look at how equivalent statistics are treated in other sports; in soccer for instance the idea that you could accurately predict a team’s place in the standings on the basis of goals scored/lost would be roundly ridiculed because there’s a general acceptance that some teams are constructed in a way which lets them win big in comparison to other, technically equal, teams.
Actually, a similar relationship DOES hold in basketball, which also a) has a shitload of games b) where opponents play each other multiple times during a season. You can find a good relationship between point differential and winning percentage.
Look, runs win games- so why would you assume runs scored/allowed over a season isn’t related to wins over a season? Or are you seriously arguing winning games 3-2 as opposed to 11-0 (or losing games 3-2 and 11-0) are different skills?
[luck != skill]
27 – Interesting. The M’s having the best OPS in close and late situations is an indicator of the luck that is skewing their record vs. pythag.
I can’t wait to see the arguments that the M’s are made up of “clutch hitters.” There is no real reason other than luck that the M’s should actually have the best OPS in close and late situations versus middle of the pack in all other situations. The fact they have hit better in these situations than their overall performance scares me for next year.
[if you don’t stand behind your ideas, don’t post them]
“Look, runs win games- so why would you assume runs scored/allowed over a season isn’t related to wins over a season? Or are you seriously arguing winning games 3-2 as opposed to 11-0 (or losing games 3-2 and 11-0) are different skills?”
I think I explicitly began my post by saying it is related (what sort of ludicrous argument would I be making if I thought runs had no link to wins!). I’m saying that there are reasons why teams win more games than their pythag would predict – that it’s not simply luck. We can sit here and say that statistically, looking across years of data, run distribution is 95% random (Dave’s argument) but that doesn’t quantify the specifics of reality. Take an extreme example – you have a rotation of 4 guys with sub-3 ERAs and one guy with an ERA over 30 and you run out the same rotation for the entire year, you’d better believe your pythag isn’t going to match your actual win-loss record. How does pythag capture that?
If you’re going to respond with “that’s a ludicrous example” then you’re correct, but the point is that the construction of a team can clearly skew pythag predictions. If it can in an extreme context then why aren’t these influences at play in slightly less ridiculous scenarios?
If you’re going to respond with “that’s a ludicrous example†then you’re correct, but the point is that the construction of a team can clearly skew pythag predictions
And my argument would be “I don’t think it’s really possible to construct a team to consistently skew pythag very much”. In other words, losing 11-0 and winning 3-2 doesn’t work over the long run. Good teams blow bad teams out and get blown out less.
If Pythag was bunk, it wouldn’t work as well as it does- and it’s worked over 100 years of baseball. Consider that this includes dead-ball, 2-1 games in the teens and 1960’s as well as the go-go-30’s and 90’s-00’s, all times with wildly different playing conditions, different pitching usage patterns, even markedly different rules… but Pythag still works (whereas Bill James had to diddle with Runs Created during the deadball era, and so on). That, to me, speaks to it’s utility as a measurement, that it actually holds true under a variety of game conditions.
30 – I don’t think the argument is that construction has *no* effect, it’s that it would be rare to have it be significant enough to rise above the luck “noise”.
You call out a rather extreme example, but let’s flip that around. When was the last time you saw an MLB team run out a starter with even a double-digit ERA for a complete season? How about a sub-.150 BA from a starter (who’s not a pitcher)?
I’m not anti-Pythag. I’m just trying to figure out how a flowing, changing dataset can possibly be used to predict anything. I think it can be used as a “should have” point, certainly. As in, the Yankees SHOULD be a better team then they have been. And I think it can be used as an indicator that a team is “hot” or “cold”. But I don’t like the idea that because the current Yankee run differential is so damn high that it means that their forward progression is so obvious.
The Yanks have had the benefit of an absurdly easy schedule against teams that are barely Major League quality which has caused a huge upsurge in their run differential. That’s a “hot team” against “bad teams”, not a “this team is going to go 40-10 the rest of the way because of how they’ve played those bad teams”.
If I’m off the mark on something, I am open to education, but I don’t like the idea of picking and choosing when to use the run differential to point out future projection. Even two months ago I agreed that the Yankees were a better team then their record indicated thanks to their run differential, and I still agree to that point. But two months ago their run differential wasn’t as severe as it is now.
On June 1st, the Yankee run differential was a mere +28 runs and their record should have been 29-23.
On July 1st, the Yankee run differential was +46 and an expected record of 44-35.
As of today it’s over +140 and their expected record should be 69-43.
Even if we could have used run differential and pythag as a predictor of future success, could we have accurately said back on June or July 1st that the Yankees SHOULD be 26 games over .500 (+/-5 wins) as of August 6?
That’s all. I like advanced metrics, analytical study of stats, and sabermetrical fun times as much as the rest of you, but when it comes to people using Pythag, I think that there are way too many opportunities for abuse to make one’s point.
And so, it’s not Pythag I guess I have a beef with, it’s the people using it.
I’ll just say, as a student who just took a few statistics courses, is that most statistics used in determining anything (whether value, depreciation, or in this case runs scored/given to predict win/loss) is a changing set of data points. The fact that Pythag has been used so thoroughly, and been seemingly very accurate is in and of itself a support for its continued use. Yes, it is not 100% correct, but what they mean by that +/-5%, is not that they have to be within their predicted value plus or minus 5%, it is rather they are 95% certain, that the true mean should fall within one standard deviation of their predicted mean.
Other than that, this year has been awesome, I’ve been able to talk trash to my friends who are dodgers, angels, red sox, yankees, and braves fans and say the mariners are coming! But, I cannot say we are a better team than any of those, except maybe the dodgers, for sure. we have glaring holes that everyone seems to recognize except the managers. Or they recognize it, and don’t think it is within their power to fix it. I firmly believe AJ should be the starting LF instead of Raul, but do I know for a fact that he will do better? No. But I have a lot of faith that he will be a considerable upgrade. We do need pitching, we need a better 1B, but there are a limited number of options. Thats why I read here, Dave, DMZ, and a lot of the posters provide clear concise arguments, using stats and a bit of gut instinct in their evaluations, which is a far cry more than most. If we make the playoffs, itll be because we somehow manaed to win where we most likely shouldn’t have. I hate that word luck. Luck smacks of disrespect and unbelief to me. Whether bad or good.
Sorry for the novel, I sorta seemed to have a need to say it all, since I cannot admit this to any of my friends with other teams.
33. I think TIF gets at where the problem lies. I agree that there is nothing wrong with saying that a team that has outperformed its Pythag has been fortunate, but I take issue with someone even implying that a team’s Pythag is proof that they cannot compete in a playoff race.
If we’re going to call into question the Mariners’ chances at the playoffs, let’s cite their questionable rotation, slap hitting DH, insistent usage of a slumping, slug footed LF along with other liabilities, and their failure to maximize the talent they have on the roster and in their system… over citing their runs scored vs runs allowed and comparing it to other teams’ runs scored vs runs allowed.
Yes, because none of those things show up in their RS/RA.
It’s a rough, entirely effective measure of a team’s true strength. I don’t get the hate, and I’ve written entire pieces on the limitations of Pythag.
35: I agree. Just because the Ms have more wins than they ‘should’ doesn’t mean that they don’t have a shot at playoffs.
If the management puts their heads on straight (unfortunately, they might be as well on as they will ever be), then the Mariners can turn the ‘luck’ wins into straight wins, and expand their run differential. If the glaring holes that can be filled, are, then I have no reason to believe that the Mariners can’t contend for a playoff spot.
On a somewhat unrelated note… [deleted, somewhat unrelated note]
I wouldn’t call it hate… maybe, as of late, frustration with over-reliance on an admittedly limited measure. It’s an entirely effective measure… after game 162, but until then, it’s measurements are still being taken with every game.
Some contributing factors to a team’s RS/RA or lack thereof can be eliminated. For example, two key products of the 16-1 Reds loss, Ryan Feierabend and Jason Davis, two guys who weren’t much better in other instances they pitched, are not parts of the team anymore. Players also improve and decline: April 07 Jeff Weaver, who directly contributed to our RA total, is not the Jeff Weaver we have today. Sexson has contributed to our lack of RS: he’ll either get better or stay benched. We can spend all day digging for other direct examples.
Granted, all 30 teams have supplanted lesser parts with greater parts, but not all lesser players are equal, nor are the greater parts created equal. It’s not a given that the M’s, Pythag-wise, stay where they are. Nor is it with the Yankees, Angels, Tigers, Indians, and so on.
Does that make the M’s top contenders? Hell no. But there are several more tangible, direct reasons for that.
I don’t get the hate
You and me both, but I suspect it’s because people don’t like hearing “Gee, you got lucky”.
I have NO PROBLEM with my team getting lucky. Consider the 2002 Angels, 2003 Marlins and 2006 Cardinals…
You can do that with any measure, though, and arguing that the Jeff Weaver we have now isn’t the Jeff Weaver we had then is just… it’s like Gillick’s pro-Franklin arguments. You’re arguing that a stat’s invalid because it’s measuring what happened, but that’s part of what happened.
And that it measures where a team is? Yeah. It’s a rough estimation of a team’s actual strength based on their performance to date. It’s no less valuable because the season’s not done, any more than a player’s OBP isn’t valid because they’re only partly way through the season.
That’s not a bad analogy, actually. OBP isn’t a perfect measure of what happens at the plate, and you could get plunked more or less than you will going forward. But it’s still a fine way to look at player effectiveness.
You know, Gomez, I don’t mean to be rude, but every possible issue you could raise with your dislike of pythag has essentially been refuted. It’s more accurate than you think, and the continual hand waving of pythag because we don’t like the results is no different than people who say Vidro’s awesome despite the lack of power because he’s hitting .300.
This may have a simple answer, I’m trying to follow this discussion even though I’m still learning the math side of all this…
I don’t understand how this statement:
“Does that make the M’s top contenders? Hell no. But there are several more tangible, direct reasons for that.”
…is incompatible with the whole RS/RA issue. So I’m having a disconnect on the disagreement here.
Aren’t all those “tangible, direct reasons” just reflected in the RS/RA?
Am I missing something? No one is saying that the RS/RA is the “reason” we’re probably not going to get the wild card, but rather a reflection of the reasons we probably won’t be able to get the wild card, right?
I guess I really just don’t understand the point of contention.
Dave, are you still mad at me after all this time?
How do my doubts of Pythag’s predictive ability become ‘hate’? Explain that to me.
That was me, actually, in describing pythag doubts. Don’t take that personally.
No, Gomez, I’m not mad at you – I’m somewhat annoyed with the constant desire of otherwise intelligent Mariner fans to ignore an obviously useful tool because they’re not happy with the results it’s providing.
Okay, fair enough.
Nobody’s saying pythag is completely useless, but instead of explaining why potential limitations don’t matter, we should have some way to account for them in my opinion.
For instance, how do we address a bullpen with three lights out relievers who pitch in crucial situations and three dismal relievers who generally pitch when the team’s behind? Good pitchers give up less runs than bad ones, good pitchers pitch in crucial situations more than bad ones, consequently the time when runs are given up can be altered in a way which isn’t a skill or completely random. A sizeable proportion of us have spent time criticising Hargrove/Maclaren’s bullpen management under the premise that it has a significant impact on the Mariners winning games, so why does that become so insignificant as to be ignored when we discuss pythag? Surely a team with 3 lights out relievers and 3 abysmal ones will outperform their pythag in comparison to a team with 6 average but collectively equal relievers – because the time when runs are given up can be controlled to a greater extent by bullpen management.
I don’t see how pointing to broad trends in run distribution negates that point. Run distribution accounts for the offence, it argues that there aren’t streaky offences who put up 12 runs one night and one run the next all the way through the year, but I don’t see where issues like the bullpen fit in. If you look at WPA for instance, you’d conclude that Putz (and other members of the bullpen) have allowed the Mariners to win a number of games. If instead of Putz pitching in crucial situations we had put Reitsma in there instead I’m sure we’d see a tangible difference in our win/loss record even if the grand total of RS/RA remained the same. Why is that not significant?
Incidentally, apologies if I’m clogging up this comment section, I realise the original post wasn’t on the merits of pythag. I completely agree with the original point, I think we have a much better chance of winning the division than getting the wildcard and I think the Yankees are a better team than the Mariners.
#39
Not sure using the 02 Angels as a team that was lucky in relation to their Pythag…. The 02 Angels may have gotten hot during the postseason but they also had the best run differential in MLB that season.
It’s way too late to bring this into the discussion given that this entry is no longer on the front page, but as a final follow up to this “If you look at WPA for instance, you’d conclude that Putz (and other members of the bullpen) have allowed the Mariners to win a number of games.” consider that out of 9 teams who have an actual win-loss record higher than their expected record, 8 of them have a bullpen which has a positive WPA and 7 of them have a bullpen with a WPA above 4 – which would correlate with being 8 games above .500. The only team who has a record above the record expected using pythag which doesn’t have a bullpen with a positive WPA are the White Sox (-1 WPA).
If Pythag simply measures luck and things like the bullpen are insignificant, then you have to admit that’s a bit of a coincidence.