Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

Let's start off with a little bit of housekeeping.

First, I discovered an error in the code that built the batting lines; this was leading to overstated numbers of extra-base hits and slugging totals. This shouldn't have affected the game odds materially (since each team's extra-base hit totals were being overstated in the same fashion), but did lead to overstated slugging percentages in the batting lines. Apologies for the mistake.

A couple of reminders of what we are and aren't doing here:

  • Unless otherwise noted, we are not adjusting for the platoon splits – a left-handed hitter will project the same versus right-handed hitting or left-handed hitting. Platoon adjustments are among the very last steps of the PECOTA process, and we're grabbing the limited forecasts used for this from an earlier stage in the PECOTAs (before age adjustments are applied.) We have for a few games worked in a platoon adjustment by hand. Since this is suboptimal, and it continues to come up, I will try and get a more robust platoon adjustment into Tuesday's updates.
  • We are using linear weights to do the conversion from batting stats to runs scored; a change in the lineup order will not affect the forecasts. Doing a proper lineup simulation would conceivably improve the model (of course, that requires the model to be done correctly, which is probably harder than doing it incorrectly). Lineup optimization is, I think, an underexplored area in sabermetrics – there are some publicly available models that are based upon historic data, but spit out fairly ludicrous lineups regularly. It's a consequence of being non-interactive – baseball managers tend to behave a certain way when they build lineups, and if you don't account for that, your results will be in error.

And to assuage your worrying – Eric will be back with you Tuesday. I'm sure you miss him as much as I do, if not moreso (since I just finished talking with him about five minutes ago, I'm betting moreso). And thanks to Eric to helping me out with this through the playoffs.

And now for the game itself. An inability to decisively finish off the Rays after a promising start has put the Rangers is a tough spot, with Cliff Lee taking the mound for the first time in Game Three. The Rangers were a bullpen meltdown away from coming into this game with a lead in the series; but the bullpen meltdown happened and now they're tied 1-1 and on the road.

To expand upon what I talked about Sunday – home field advantage in a seven-game playoff series is an odd duck. Here's how it breaks down by series length:

  • Four games – No home field advantage
  • Five games – Lower seed has home field advantage
  • Six games – No home field advantage
  • Seven games – Higher seed has home field advantage

Despite an August spent in a half-hearted pursuit of the AL East crown (and thus presumably a better playoff seed), the Yankees will end up with the real home-field advantage in this series if they can take two out of three here at Yankee Stadium. And they greatly increase their chances if they can take this game from the Rangers. (In spite of the hand-wringing about A.J. Burnett in Game Four, the Yankees have a pretty good chance to win a game at home against Tommy Hunter.)

And it might surprise some people, but running the numbers through PECOTA suggests the Yankees are (slight) favorites to win Monday night, at 53 percent. Yes, Lee is a dominating pitcher, but Andy Pettitte isn't a slouch. The Yankees also feature a significantly better lineup (which is how they scored more runs in the regular season, despite playing in a less-friendly park to hitters) and have the advantage of playing at home.

A look at the lineups versus the starting pitchers – first, the Rangers versus Pettitte:

 

Hitter

AVG

OBP

SLG

Elvis Andrus

0.238

0.306

0.360

Michael Young

0.291

0.328

0.473

Josh Hamilton

0.296

0.339

0.547

Vladimir Guerrero

0.302

0.339

0.529

Nelson Cruz

0.255

0.308

0.490

Ian Kinsler

0.294

0.337

0.520

Jeff Francoeur

0.263

0.293

0.439

Jorge Cantu

0.272

0.310

0.459

Bengie Molina

0.290

0.306

0.476

And the Yankees versus Lee:

 

Hitter

AVG

OBP

SLG

Derek Jeter

0.289

0.332

0.455

Nick Swisher

0.273

0.326

0.515

Mark Teixeira

0.289

0.344

0.548

Alex Rodriguez

0.288

0.342

0.563

Robinson Cano

0.311

0.331

0.527

Marcus Thames

0.257

0.292

0.506

Jorge Posada

0.283

0.334

0.506

Curtis Granderson

0.272

0.314

0.498

Brett Gardner

0.246

0.298

0.386

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
BeplerP
10/19
To me, all these projected stats seem inflated. Why is this? Small sample size effects? Thanks for the work, but I'm just as puzzled as before. And not just because I knowe the Rangers won this game 8-0! Regards,
doctawojo
10/20
Off topic, sorta, but I'm curious what you mean by: "there are some publicly available models that are based upon historic data, but spit out fairly ludicrous lineups regularly". Specifically, the "ludicrous" part. Do you mean lineups that no reasonable manager would ever use? Or lineups that are simply _wrong_? (E.g. third-best hitter on the team hitting eighth or something like that.)