Barry Zito did something strange Wednesday night against the Cleveland Indians. He pitched well. He didn’t walk a hitter over 6.2 innings. He had his longest start of the year. And he won a baseball game.
His final line of the night was: 6.2 IP, 4 H, 1 R, 1 ER, 0 BB, 4 SO
All good things. Zito appeared to be throwing a tick harder than usual, he touched 87mph in the 6th inning and for him, that’s like dialing it all the way up to 97. The result of a lower arm slot? Maybe, who knows, I’m not a mechanics expert. This was also his first start of the year where he did not issue a walk. Zito’s control has been a major issue for him this year. The start last night lowered his BB% to a still scary 12.6% but it’s no longer in the 13′s like it was.
Ultimately, though while encouraged by Zito’s start, I think it raises just as many questions with me.
Questions such as:
His velocity, was it really the result of different mechanics or something else? Is it sustainable? Or will it come and go?
How much did facing Cleveland, a team that’s struggled offensively all year, help Zito? Any night you can get out major league hitters is a good night but there’s a big difference between the Indians (EqA of .250) and Barry Zito’s next opponent, the Cubs (EqA of .276). The Cubs have the 2nd best offense in the game right now, only behind Texas.
Did his new arm slot help his control? Zito was walked as many as 6 hitters in a game this year and his control is at a career worst.
I think the best thing we can do is to not get overly excited and believe that “Zito has figured it out” because it’s just one game and anything can happen in one game, but we can take the positives as small gains and hope that the next start we’ll see the same positives again. This was definitely the best game that Zito has thrown this year. I decided to plot his WPA Scores after this game and here are my results.
If you’re unfamiliar with Win Probability Added, or WPA, it’s a basic way to figure out how much a player contributed to a win. WPA is more descriptive than predictive, so we can’t assume because Zito did well in this start that he’ll do well in the next start, but it’s a fun stat to use when looking at past games. Here’s a pretty good description of why you should care about WPA from Fangraphs, one of the best sites out there.
WPA takes into account the importance of each situation in the game. A walk off home run is going to be weighted more then a home run in a game that has already gotten out of hand. This makes it a great tool for determining how valuable a player was to his team’s win total.
Because WPA is based on the context of the situation in which a play occurs, the scores can get tilted a little bit for players that constantly appear in high leveraged situations (such as a closer that’s always used in close high leverage games) but overall, it’s a pretty handy and fun stat to look at. A WPA of zero means that the player was purely “average” in contributing towards a win, he neither helped or hurt his team.
On my plot, negative WPA score’s are indicated by the white bars and positive WPA scores are indicated by the orange bars. Also, the number above each bar indicates the game. So, Zito’s last game against Cleveland, and his best start of the year by WPA, was his 16th start.
Here’s the data that went into the plot.
GS Date Team Opp WPA 1 31-Mar Giants @Dodgers -0.216 2 06-Apr Giants @Brewers -0.187 3 11-Apr Giants Cardinals -0.214 4 16-Apr Giants Diamondbacks -0.169 5 22-Apr Giants @Diamondbacks -0.371 6 27-Apr Giants Reds -0.429 7 07-May Giants @Pirates 0.003 8 12-May Giants Astros -0.022 9 17-May Giants White Sox 0.017 10 23-May Giants @Marlins 0.108 11 29-May Giants @Diamondbacks -0.079 12 03-Jun Giants Mets -0.236 13 08-Jun Giants @Nationals -0.042 14 13-Jun Giants Athletics -0.191 15 18-Jun Giants Tigers -0.322 16 25-Jun Giants @Indians 0.355
As you can see by the plot, Zito’s first 6 starts of the year were disastrous. The worst being the 6th against the Reds in which Zito’s WPA hit a season low of -42.9%. In that game, he pitched 3 innings and gave up 8 runs. Zito then went to the “bullpen” and came back against Pittsburgh to eek out a positive WPA. That’s his 7th start that barely charted on the plot because the number was so small. Zito then pitched his 8th start against the Astros and eeked out another negative score of -2.2%.
As you can see, after Zito came back from his bullpen exodus he wasn’t really a better pitcher, just not as terrible as he was earlier in the season. He bordered more on average than anything else. His 7th ,8th , and 9th starts all hovered right around 0 WPA. Zito finally got his first win against the Marlins on May 23rd, and until that point, it was his highest WPA scored of +10.8%. That’s until his game last night when Zito scored a +35.5% by WPA, his highest score of the season by far.
Like I said above, I’m not sure if Zito has necessarily turned a corner, but you can’t help but root for the guy despite his large contract. He seems genuinely nice and it was nice to hear him talk about getting a win on the post-game last night rather than explaining why he melted down. Again, I think it’s important that we don’t put too much into his start last night other than appreciate it for what it was, his best game of the year.
He’ll get a real test in his next start against the Cubbies.
—
Edit: I decided to add plots for WPA scores for Lincecum, Cain, and Sanchez, just as a comparison point.
I included Lincecum’s funky 4 inning relief appearance in his plot.
Very nice stats! Very interesting.
But picking a nit, Zito WAS “better” after his bullpen “break”, just not good, which is the spot-on point I think you were trying to make, as what use would he be for the money if he’s only average or near average. We need above average performance.
And that’s what he delivered yesterday. That .355 WPA in yesterday’s start would be among the best for Lincecum and Sanchez, and is better than what Cain has done this season, so that is a very good performance, a pretty positive sign, though not conclusive as you aptly noted.
I would also note that another key sign is his velocity returning to the high 80′s that many believe he needs to throw at to be effective. He had not reached that once his whole time as a Giant, from what I can recall. Yes, maybe he won’t repeat it, but I consider this a breakthrough.
Plus, from his quotes after the game, he finally seemed to hit bottom mentally and emotionally with the previous start, and might be in the right state of mind now to not think so much on the mound and, like Nike states, “Just Do It”.
Do you have the WPA plot for Zito’s 2007 and perhaps 2006? I think it would be very interesting to see how this last start stacks among all his Giants starts plus his starts in his last season with the A’s, which is when some people say that he started a steeper downturn in peripherals that caused them to predict that he would not do well going forward.
OGC,
>> But picking a nit, Zito WAS “better” after his bullpen “break”, just not good, which is the spot-on point I think you were trying to make, as what use would he be for the money if he’s only average or near average. We need above average performance.
I was saying that better is kinda relative when you’re stinking up the joint as badly as Zito was early in this season. In fact, I heard much hub-bub about how Zito had figured it out or how he was going to go on a roll. Which in fact when he came back, he still had control problems but wasn’t giving up 8 runs a game. Thats what I mean.
>> I would also note that another key sign is his velocity returning to the high 80’s that many believe he needs to throw at to be effective. He had not reached that once his whole time as a Giant, from what I can recall. Yes, maybe he won’t repeat it, but I consider this a breakthrough.
The velocity question is interesting because it was a nice sign but his velocity has been sketchy since he joined the Giants. If the lower arm slot made some mechanical change, that’d be great. But like most things, I need to see him do it again a few times. I’ve got the feeling it might come and go.
For fun, I did plots of Zito from 05-07. Two years with the A’s and his first year with the Giants.
His total WPA for ’05 is +0.40
Total WPA for ’06 is +2.08
Total WPA for ’07 is 0.04 (career low)
His total WPA for this year is -2.00 (The first time he’s ever posted a negative WPA and would be a career low if his season ended today).
By WPA, his peek years were 2000-2003, he had down years in 2004-2005, a boost in 2006, and drastic lows in 2007-2008.
Do you ever look at Game Scores? I’d bet you’re familiar with it but in case not it’s a Bill James stat that uses a VERY BASIC formula to measure overall pitching performance. Though imperfect, the stat is elegant, and tells a meaningful story. Baseball-Reference keeps track of the stat in its year by year gamelogs.
Anyway, I was wondering what a line graph of Zito’s average GS would look like over, say, his past 4 or 5 seasons. Specifically, I wonder what kind of precipitous decline there would be, and if that decline would begin before or after 2007.
I’d do it myself but a) I’m lazy and b) you seem to be a wizard with whatever graphing app you use.
Or not. Though, it may be something to look into if you ever get bored.
Mr Lomez,
Funny that you mention Game Scores because after working on these plots today, I was thinking that GS’s would be interesting to look at in a graph form.
And despite it’s flaws, it might be a better indicator of individual performance since WPA is based in the context of when a player performs.
Thanks for the feedback, I think I’ll plot some of them tomorrow or over the weekend. It would be easy to get them since BB-Ref’s PI tool is awesome.
[...] my post on WPA Plots — which you can find here — I had some interest in plotting Game Scores. Because I love visual aids, I decided to plot [...]