2017 Projections Time
Sorry for the posting drought; I’d say it won’t happen again, but you know by now that I can’t promise that. It’s that quiet-before-the-storm portion of the off-season, where we get furtive glimpses of pitchers stretching, or dudes in t-shirts swinging bats. The big moves have mostly all happened now (though there are some surprisingly big names still waiting for a call), and thus we know the basic contours of each club in the AL West.
I ran a similar post to this one exactly one year ago, and a few things stand out about the view from February, 2016. First, beyond the win/loss estimates, the projections biggest “miss” was the run environment of 2016. The first thing that jumps off the page about the 2017 projections is that the runs scored/allowed numbers are all bigger – we’d all become somewhat accustomed to the depressed scoring environment of 2010-2014 or so, so it’s not a shock that the projections came in low. The average AL club scored just 677 runs in 2014. Last year, that figure was 731. The key driver was the explosion in home runs, a fact that won’t come as a shock to anyone who watched the 2016 M’s. This is probably a big reason why the parity that all the projections saw never really materialized – the AL schedule wasn’t a series of 3-2 and 4-3 games, so teams that excelled at clubbing the ball differentiated themselves a bit from the pack. That was a big reason the M’s not only beat the low estimates of their 2016 win total, but overcame some truly awful defense and baserunning – two factors that would’ve been more important in a low scoring environment.
The second thing, and again, this may be in part a product of more scoring, is that parity’s gone. Last year, the projections had four of the five AL West teams within 5 wins of each other. Cairo thought that’d be between 76-81 or so, while Davenport thought they’d be bunched between 83-87, but fundamentally, the division (and the league as a whole) was tightly bunched. This, far more than the individual team win numbers, turned out to be wrong. As I mentioned during the season, the systems essentially “got” the M’s, with Fangraphs 84-win projection two wins low, and Davenport’s 87 one win high. The “miss” was what those win totals would mean for their playoff odds. 86 wins in a league where no one won 90 would be one thing, but as it happened, 86 wins was only the 7th-best record in the league, and resulted in another year without a playoff appearance. That big clump of teams clustered around .500 is gone in 2017, with the Astros projected to win at least 90 games in nearly every system, and 97 in Clay Davenport’s. Fangraphs has 3 90 win-or-better clubs, in Houston, Boston and Cleveland. PECOTA has the same three, albeit in a different order. That also means there are more “bad” teams, with clubs like Baltimore and Kansas City tumbling below 80 wins.
Sticking with the AL West, here are how Fangraphs, Baseball Prospectus’ PECOTA, Clay Davenport’s projections and the Vegas over/under line shake out for 2017:
| FG | PEC | DAV | Vegas | |
|---|---|---|---|---|
| Mariners | 83 | 87 | 89 | 85.5 |
| Astros | 90 | 94 | 97 | 87.5 |
| Angels | 84 | 78 | 80 | 76.5 |
| Athletics | 77 | 75 | 79 | 66.5 |
| Rangers | 84 | 84 | 84 | 86.5 |
(FG=Fangraphs, PEC=PECOTA, DAV=Clay Davenport and Vegas=over/under betting line)
Unlike in past years, there are some considerable disagreements concerning the order of teams. The Astros are always first, but BP and Davenport both have the M’s as the clear runner-up, while Fangraphs has them 4th, barely behind Texas and Anaheim. The betting line follows PECOTA and Davenport, perhaps weighting each team’s 2016 record a bit more heavily*. PECOTA, Davenport and the CAIRO system are all in broad agreement about the M’s offense, and while there’s a bit more disagreement, on the run-prevention unit as well. In each of these systems, the M’s produce between about 60-75 runs. Fangraphs is the most pessimistic on both sides of the ball, producing a run differential of just 21, which accounts for their low projected win total.
The degree of agreement about the offense (Fangraphs aside) is striking and admittedly exciting. The M’s look better to me vis a vis their peers as a result from having dove into this than I’d thought. Some of this is the result of improved projections for the M’s core players like Nelson Cruz (whose 3-year run is remarkable, and serves as a counterweight to the aging penalties the projections will assess him) and Kyle Seager, whose breakout year boosts his projected slugging percentage. The team’s starting pitching is more of a question mark, but PECOTA gives them a big assist from their defenders. PECOTA’s team fielding runs above average metric sees the M’s as far and away the class of MLB. They’re projected to save 46 runs above average, and no other team is even in the 30s. The M’s are *18 runs* better than the 2nd best team (the Dodgers), and 26 runs above 2nd place in the division (the Astros). That’s helpful, but even with Jarrod Dyson in the fold, I’m going to take the under on that. Still, the improved defense is one reason why the projections don’t see the back of the rotation and the 6-8th starters as a black hole – they’ll give up fewer runs than they ought to.
The M’s depth is a strength overall, I’d say, with Danny Valencia, Guillermo Heredia and team-favorite Andrew Moore all looking surprisingly solid. The same’s true of some of the rookies the M’s will depend upon if they’re going to challenge Houston, namely Mitch Haniger and Dan Vogelbach. While the playing-time projections are really hard for the rotation (Iwakuma and Paxton could throw 100IP each or 200IP each, and it wouldn’t really shock anyone), it’s encouraging that they have a bit of depth in Moore, Dillon Overton, Chris Heston and Rob Whalen. None of those guys is great, but given the M’s struggles in avoiding line-up or rotation black holes, it’s nice to have options in case Yovani Gallardo face-plants again or if Paxton’s injury problems return.
* This probably accounts for what looks like, on paper, as the biggest gap between expected team strength and the betting line. Oakland’s (admittedly early) over/under total of 66.5 is 12.5 wins below their Davenport win total and over 8 wins below their most pessimistic projection.
Projections this early are just guess work. No matter the paper strength or weaknesses we all know that luck, injuries and unexpected improvements and failures will impact teams seasonal output. If the game were predictable it wouldn’t be nearly as much fun to watch.
You’re right Maq, let’s all just base our predictions on hopes and prayers
Thank you, Marc! Baseball is back; Mariners in uniform are doing baseball again. GM Jerry has raised the floor significantly in his short tenure. I’d wager whoever is the 26th, 27th, 28th, Mariner when Spring Training ends will be a better player than the 25th on half of MLB teams this year.
Thank you for the article, Marc!
Looks like Oakland would be an interesting – and potentially lucrative – bet. But, on the other hand, the people in Vegas are so obscenely wealthy precisely because they are good at setting betting lines.
Wow, a 7 win difference between Fangraphs and Davenport for the Astros. 97 seems ridiculously high for a projection.