Make It Official! Twins 3, White Sox 2: Clutch Kepler, Steely Thielbar Slam Door on Sox

[Jocko87]

6 minutes ago, CCHOF5yearstoolate said:

Leverage index isn't a factor in WPA. It's analogous to the concept that the change in win probability gets larger and larger towards the end of games in that it tries to quantify those swings, but it's a separate metric. 

Also there's a typo in the article, Thielbar actually had a 0.206 WPA per Fangraphs. 

When Thielbar entered the game, the Twins had a 79.4% Win Probability. When he exited the game, the Twins had a 100% win probability.

Good call out on leverage index, I was assuming on that one before I dug into the math.  I was assuming because it should obviously be part of the equation, how else would they know what the big moment is?  Just later in the game isn't congruent with sending the closer out against the heart of the order in the 7-8 inning, also a darling (and correct) move of the nerds and seamheads alike.  It's the correct high leverage moment.

Why would Thielbar get all 20.6% of the WPA?  One of his outs had a .630 xBa and it was one of two good plays Kepler made in the inning.  Because he gets credit for whatever his defense does, which ruins the stat for pitchers, IMO.  Lord help the Angels pitchers going to arb if they want to talk WPA.

If roughly 20% is available for a one run lead a good closer should run away with the title, but they don't.

From Fangraphs-"A fun way to think of WPA is as a storytelling statistic."  If your story doesn't include defense it's going to have a large blindspot when Buxton robs a homerun for Duran.

[CCHOF5yearstoolate]

21 minutes ago, Jocko87 said:

Good call out on leverage index, I was assuming on that one before I dug into the math.  I was assuming because it should obviously be part of the equation, how else would they know what the big moment is?  Just later in the game isn't congruent with sending the closer out against the heart of the order in the 7-8 inning, also a darling (and correct) move of the nerds and seamheads alike.  It's the correct high leverage moment.

Why would Thielbar get all 20.6% of the WPA?  One of his outs had a .630 xBa and it was one of two good plays Kepler made in the inning.  Because he gets credit for whatever his defense does, which ruins the stat for pitchers, IMO.  Lord help the Angels pitchers going to arb if they want to talk WPA.

If roughly 20% is available for a one run lead a good closer should run away with the title, but they don't.

From Fangraphs-"A fun way to think of WPA is as a storytelling statistic."  If your story doesn't include defense it's going to have a large blindspot when Buxton robs a homerun for Duran.

I'm 90% sure even leverage index doesn't take into account who is batting. You're right in that we know it's more valuable to shutdown the heart of a lineup in the 8th than it is to pitch against 7-8-9 in the 9th, but I don't know if any of these kinds of metrics have caught up to that yet.

Yeah WPA in my opinion is an OK estimator for pitcher performance in the aggregate but it breaks down quickly under any scrutiny. Similar to how ERA can be deceptive depending on the quality of the defense, it's a descriptive of strictly what happened but it's not all that helpful when you want to evaluate an individual player's performance. 

[JD-TWINS]

10 hours ago, rv78 said:

It's great to beat up on the two worst teams in the AL, but that only means you could still be the 3rd worst team in the AL. I'll start believing when they start beating good teams. 

Way to be upbeat!