Hey there, everybody! My name is Hokius and I've been around here for a couple years, I don't post a whole lot but I'm always reading. I thought I'd try my hand at a little bit of sports blogging this year, Win Probability and Win Probability Added are among my favorite stats as a super amateur math/stats nerd. They look like fancy stats, but they're honestly really easy to understand. Let me give a quick breakdown for those who haven't looked at them before.
Win Probability/Win Expectancy (WP/WE) can be calculated based on any set of circumstances on the field. (Here's a really neat calculator I found in a quick google search, try it out! http://gregstoll.dyndns.org/~gregstoll/baseball/stats.html#V.0.1.0.1) and basically what it tells you is the probability of one team winning over the other given the specific set of circumstances. One of the more famous examples for us Royals fans is Game 4 of the ALDS last year (2015). The Royals had a 3.2% WP after Marwin Gonzalez struck out to end the 7th. By the time Alcides Escobar struck out to end the top of the 8th, the Royals had a 70.8% WP. That's a HUGE swing. And of course, the Royals ended up winning that game; don't you wish you'd put some money on it after the 7th, now? Anyway, so it's pretty easy to understand. I don't honestly know how to calculate it, if I enjoy writing these I'll probably see if I can't figure it out eventually, but I suspect the math might be a bit over my head.
Win Probablity Added (WPA) you can probably guess from the name and an understanding of the previous stat. It's simply the difference in Win Probability before a given moment and after. And that's what I want to look at a bit today, in regard to last night's game.
Royals 4 - Mets 3
(Link to FanGraphs BoxScore complete with WPA Info: http://www.fangraphs.com/boxscore.aspx?date=2016-04-03&team=Royals&dh=0&season=2016#plays)
I want to look at a couple different things: when the Royals win, I'll look at the moment it most looked like they were going to lose; and vice versa if they lose. And I want to look at the 3 plays that caused the biggest swings in any direction and see if we can't figure out why it meant so much.
Last night, the Royals won so let's look at the moment they were least likely to lose:
Uhhhh...welll...that would be the beginning of the game. Guess it helps when you take an early lead as the home team and never actually relinquish it. the next lowest moment, after the first inning, is when Asdrubal Cabrera singled to right in the second inning, which dropped the Royals down to a 60.6% chance of winning (which I would still take most nights, if you gave it to me at the start). This single happened with 2 out and put men at first and third for Travis D'arnaud. The reason it was such a 'low' moment for the Royals seems pretty obvious, there was a man in scoring position and the Mets were only down by 1. According to that calculator I linked earlier, the odds actually would have favored the Mets at this point had there been no outs (Now the tying run can score even on an out, and the go ahead run can be driven in without more than 1 single in a row) but it swings almost 7% per out after that. So we see already that how many outs are in the innings makes a pretty huge difference in the WP. The calculator I'm using doesn't take into account the quality of the players pitching, hitting or on the bases. I don't know if the FG one does, but I know they differ by about .4% in the actual scenario, for whatever that's worth.
OK, so the Royals were always pretty well favored to win this game. Let's look at the 3 biggest WPAs in the game:
When: 2 outs in the top of the 9th.
Who: Wade Davis pitching, Yoenis Cespedes batting.
What: Cespedes strikes out with men at 1st and 3rd.
How: The at bat took 8 pitches. Wade started him off with a fastball pretty much down the middle that Cespedes took. Is there a scouting report on Cespedes that said he'd take in that situation, or was Wade just guessing? Then Wade froze him with a curveball on the inner half. Davis then tried to get him in Cespedes' known weak spot: high heat. Cespedes got a piece and fouled it off. Davis and Perez changed his eye level with a fastball outside, Davis missed with another fastball that was supposed to be outside and was fouled off from the inner half. Gave him another curve that he just got a piece of, then a curve in the dirt. Finally Davis went back outside with the fastball and Cespedes flailed at it helplessly.
Why: This is another obvious one, this is a big play because the tying run was again at third base with 2 outs, for a Royals Win Probability of 83.8% When Cespedes struck out, the Win Expectancy for the Royals jumped to 100% because the game was over and they still had the lead.
When: 1 out in the top of the 9th.
Who: Wade Davis pitching, Curtis Granderson batting.
What: Granderson singles into right, driving Alejandro De Aza (who had pinch hit for Juan Lagares and reached on a Fielder's Choice) to third base.
How: Wade and Curtis had a 4 pitch battle, Wade got ahead quickly with a nice curve ball on the outside corner followed by a fastball on the inner half that Granderson pulled foul. Aaron Boone, on the ESPN broadcast, announced that Davis needed to make Granderson expand his zone with pitches that weren't strikes. Wade threw 2 straight fastballs outside. The first one was fouled off to the 3B side. The last one went up the middle on the ground. Not outside enough.
Why: This one is only slightly more difficult to understand than the last one we looked at, before this AB the tying run was at 1B requiring an extra base hit to drive him home. After, the tying run could be scored without the benefit of a hit, and it no longer required a home run to put the Mets in the lead, a double would probably do it.
When: Top of the 9th, 1 out.
Who: Wade Davis pitching, David Wright batting.
What: David Wright Strikes out with men at 1st and 3rd.
How: Wade Davis just flat over-powered Wright. The ESPN announcers noted that Wright had been behind fast balls the entire game, earlier in the game Jessica Mendoza also outlined how Wright's nagging injuries are still slowing him down, even at the beginning of the season with his pre-game routine being changed to allow for more warm up and less actual activity in the form of fielding and hitting practice. Davis gave him a fastball down the middle which Wright was behind. Another which Wright fouled off into the seats on the 1B side. And then a front-door cutter which Wright could only stare at helplessly. For what it's worth, Wright thought the pitch was inside. ESPN's pitch tracker thought it was solidly in the black. Either way, Wright was probably not going to get a hit off Wade Davis last night.
Why: If you were paying attention to the previous entry, you probably already know the answer to this: Granderson's result was worth so much because it put a man at third where a productive out could tie the game. Wright's result was worth so much because it was not a productive out and removed the possibility of tying the game without a hit. The only way he could have done worse would have been to ground into a double play and end the game.
I guess it's a good day when the most disappointing thing about the Royals game was that there wasn't a large variety of results leading to huge WPAs. Hopefully I'll be back on Wednesday with another post about a Royals win with more varied WPA results! Thanks for reading, feel free to leave any comments or suggestions! Especially if you can tell me why the embedded Win Expectancy chart I can see in the editor is showing as a large blank space in the actual post. :)