Are the Cubs Vulnerable to Strike Throwers?
I might need to prime you for this one first. I don’t know if this is something people are actually talking about, or if I just made this theory up. But it sounds like it could make sense, so let’s go forward. In Game 3, the Indians are throwing Josh Tomlin against Kyle Hendricks and the Cubs, in Chicago. That doesn’t seem to bode very well for the visitors. The Indians were already the underdog, and now they’re losing the DH and the home-field advantage. Our scoreboard page puts the Cubs’ odds of winning at 64%. You could argue a few points in either direction but the Cubs are supposed to win this game.
However, let’s theorize! How might you make yourself feel better about the Indians’ shot? We know Tomlin is by no means overpowering. He’s not a strikeout pitcher. What he is is a control pitcher, and in that regard he’s one of the best. He led the American League in strike rate, among qualified pitchers. He had baseball’s lowest walk rate, among qualified pitchers. Tomlin is forever around the zone, and, the Cubs offense just finished with the highest walk rate in baseball. Being disciplined about the zone has been a big part of the Cubs’ offensive equation. Doesn’t it make sense that the Cubs could struggle against strike-throwers, who consistently get ahead? Could Tomlin turn the Cubs’ discipline against them?
Pretty interesting theory, right? Yeah! No. Sorry, but, nope. Tomlin isn’t poised to turn the Cubs’ walk rate into a weakness. Sorry if I just spoiled the rest of the article, but if Tomlin is going to succeed, he’s going to have to do it simply by executing almost perfectly, just like everyone else.
We’ve run some similar studies before. Recently, I tried to take a look at whether the Cubs struggled to an unusual degree against great pitchers. We’ve also examined whether the Cubs struggled against power pitchers, or strikeout pitchers. This is another little study powered by the Baseball-Reference Play Index. I looked at the numbers for every pitcher who faced the Cubs this season. I looked at their numbers against the Cubs, and I looked at their numbers overall, against everyone. I separated the pitchers into four groups, of 77 pitchers each, ordering them by their overall walk rates. In the table below, Group 1 includes the 77 pitchers with the lowest season walk rates. The Group 4 pitchers had the highest season walk rates.
Here’s everything:
| Group | PA | Cubs BB% | Group BB% | Ratio% | Cubs OPS | Group OPS | Ratio% |
|---|---|---|---|---|---|---|---|
| 1 | 1522 | 6.2% | 5.3% | 118% | 0.745 | 0.718 | 104% |
| 2 | 1715 | 9.3% | 7.6% | 123% | 0.789 | 0.760 | 104% |
| 3 | 1869 | 11.4% | 9.7% | 118% | 0.749 | 0.753 | 99% |
| 4 | 1229 | 15.3% | 13.0% | 118% | 0.828 | 0.802 | 103% |
Obviously, this year, the Cubs walked the least against the stingiest pitchers. The Cubs’ walk rates go in the same direction as the pitchers’ overall walk rates. But the ratios are pretty consistent, with the Cubs out-walking the rest of the league by roughly 20%. It was the same against the best control pitchers and the worst control pitchers. And the next columns are even more important. By OPS, against the best control pitchers, the Cubs offense was 4% better than average. Against the worst control pitchers, it was 3% better than average. There’s maybe a little blip there in Group 3, but I wouldn’t read too much into it. Once again, there’s consistency here. The Cubs do better than average, whether the pitcher is a strike-thrower or a strike-wanter.
I’m sure it can’t be that surprising. It’s very possible this is looking into a theory you’d never heard before this very post. Now, it’s true I could’ve done this with greater rigor. I didn’t control for the identities of the Cubs’ hitters, and maybe they weren’t distributed evenly. I’m extremely skeptical that would be true, though, and I just don’t see any signs of a real effect. The thing about the Cubs is, they can hurt you in different ways. They’ll draw walks if you’ll provide them. They’ll hit strikes if you’ll provide them. This is typical of really good lineups. They can handle strikes and balls, and all this means is the Cubs might be a little extra aggressive going after first or second pitches.
The Cubs didn’t have trouble this year against Bartolo Colon, or Jameson Taillon, or Mike Leake. They struggled against, say, Johnny Cueto, but Cueto is something beyond just a control artist. Tomlin’s whole game plan is strikes, and that shouldn’t give him any particular advantage. Every at-bat is going to be difficult. And Tomlin almost certainly shouldn’t see the sixth inning. Perhaps not even the fifth.
I mentioned the other day that, according to our splits — which go back to 2002 — Tomlin has the biggest difference on record in performance with the bases empty, and with runners on. With the bases empty, he’s allowed a .303 wOBA. With the bases not empty, that’s shot up to .367. Nobody’s split has been worse, so that could be one key for Tomlin — keep the leadoff guy away from the basepaths. It would also suit the Indians to prevent Tomlin from even attempting a third trip through the order. I don’t think Terry Francona would dare to allow that, not in the World Series, not with a fresh bullpen. Tomlin’s outing should be limited, and he’ll want to keep people off base. That doesn’t even qualify as advice. That’s just saying something obvious.
The Indians will be underdogs in Game 3, just as they’re underdogs in the series. I don’t think there is any special key for Josh Tomlin; he just needs to hit his spots as often as possible, and hope. He’s not a bad pitcher, so that’s something. And he definitely impressed against the Blue Jays. This is another chance for the Indians to try to steal a win in advance of another turn for Corey Kluber. Do that and the momentum shifts. I just wouldn’t be counting on them to do that.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
The wind is expected to be blowing 20-25 mph out to center at Wrigley tonight. For reference, the 8 home run game in last year’s NLDS had 17 mph winds out to right. Could be a very long, short night for the pitchers.
AND Tomlin gave up a ton of HRs this year.
good gravy
If this ends up like the ’03 Cubs/Cards game that had crazy wind blowing out (that ended up not counting because so many runs were being scored they couldn’t get it in before the storm system arrived to wipe out the game – it was 11-9 after 4 with 7 homers) this could be a pretty amusingly ridiculous night of baseball. The predicted wind speed is comparable.
The anti-game 2
Thanks for raising such an interesting question. I took a different approach to the study, and got quite different results — suggesting that the Cubs do have trouble with strike throwers, relative to other pitchers.
My method: First off, to get a larger sample, I used SP vs. Cubs for the last 2 years. I think this is reasonable, since 75% of their PAs in that span came from batters who were Cubs in both years. And the Cubs led the NL in walks drawn both years.
Then I broke those pitchers into 3 equal groups, based on their BB% over the last 3 years combined, and excluding anyone with less than 20 starts. That left a pool of 127 starters. Each of the 3 groups totaled between 85 and 98 starts vs. the Cubs in 2015-16.
The group totals:
Low BB (5.6% BB) — Overall 3.88 RA/9 … 3.80 vs. Cubs
Med. BB (7.6% BB) — Overall 4.62 RA/9 … 5.25 vs. Cubs
High BB (9.3% BB) — Overall 4.63 RA/9 … 5.18 vs. Cubs
Both the medium and high BB pitchers yielded about 0.6 more RA/9 to the Cubs than overall. But the low-BB pitchers fared a tiny bit better vs. the Cubs than overall.
FWIW, Tomlin has the 2nd-lowest BB% of the 236 pitchers with 20+ starts since 2014.
Interesting. Doesn’t really show something much different from Jeff. Especially if you consider changing out Castro for Zobrist.
It shows something completely different. Jeff’s study shows the cubs are just as good against’s low bb pitchers, john’s shows that the cubs were below league average against low bb pitchers and well above average against others.
Yes, 75% of the overall batters are the same. But among those, Bryant and Russell and Baez have all taken huge strides forward. And tonight’s lineup with have Zobrist, Contreras, and probably Heyward. The only ones who are really the same are Fowler and Rizzo.
I don’t think generally forward or backward is really the issue here because we’re talking about productivity in one situation relative to productivity in another situation. What’s more significant i would think would be if it’s simply the same hitters.
You really don’t think young players approaching peak is relative?
My only issue with using last year is that Heyward wasn’t on the team and he doesn’t strike out all that much compared to who he replaced (Soler/Schwarber). Then we saw Bryant and Russell cut their K rates fairly substantially this year from last.
6The relevant differences in the studies would seem to be that yours was based on ra/9 while jeff’s was based on ops, and yours was based on data from 2015-6 whereas jeff’s was only from 2016. Either ra/9 and ops for some reason give very different results in this respect or 2015 is very different from 2016 in this respect.
While the same players got those at bats, they also made improvements. See bryant, soler, baez, Russell’s k rates. Fowler was leaps and bounds better. They had no regression. Heyward swapped in being the worst hitter instead of castro. Contreras basically was schwarber. Then add zobrist as a cherry on top. Just a much better team imo.
John, can you break down the data into 2015 and 2016?
Soler- 2015 7.9 bb% 30.00 k% 96 wRC+
2016 11.7 bb% 25% k% 106 wRC+
Russell 2015 8.0 bb% 28.5k% 91 wRC+
2016 9.2 bb% 22.6 k% 94 wRC+
Bryant 2015 11.8bb% 30.6k% 136 wRC+
2016 10.7bb% 22 k% 149 wRC+
I’m on my phone so no more of this. Baez shaved off 5% of the k rate as well. Ross and Montero flip flopped for title of average hitting part time catcher.
Contreras was 6% worse than schwarber by wRC+ in a few more pa but also k’d 5% less. Rizzo was literally same same. Fowler 19% better by wRC+. La Stella got more plate appearances instead of jonathon Herrera. Plus like I said zobrist is just a good hitter in all facets.
Let’s hear from john autin if the pattern was the same in 2015 and 2016. Hopefully he’ll answer. I meant to put the above question in reply to him, not you.
JFTG, I will show the data for 2016 only, with the caveat that the samples become quite small. To combat that, I widened the pitcher pool to those with at least 15 starts over the past 3 years (instead of 20 as before).
87 of those pitchers started against the Cubs this year. Broken into 3 equal groups by overall BB%, here are the RA/9 figures overall and vs. Cubs:
Low BB (5.5% BB) — Overall 3.95 RA/9 … 4.81 vs. Cubs
Med. BB (7.4% BB) — Overall 4.30 RA/9 … 4.82 vs. Cubs
High BB (9.4% BB) — Overall 4.73 RA/9 … 5.37 vs. Cubs
So for 2016 only, we do see the Cubs beating up everyone about equally.
But again, we’re now looking at groups of between 40 and 49 starts. And about 1/4 of the pool had less than 200 IP over the prior 3 years, with 2 of them under 100 IP — so we don’t even know the true characteristics of some of these pitchers.
I’m not sure that *any* one-year study of team vs. pitcher type would be statistically significant.
The problems with multi year studies are equally distributed plate appearances and non prime age players however. Older players should regress while young players improve and the cubs are heavily weighted with the young players, a lot not even 25 yet.
2015 cubs opening day lineup was Mike olt, Davis ross, tommy la stella, Starlin castro, Chris coghlan, soler, Fowler and Rizzo. So yeah…
2015 Cubs opening day lineup is super deceptive, since it is pretty much the only lineup Kris Bryant wasn’t in.
So the stats show that the cubs were vulnerable to low bb pitchers in 2015 but not 2016? My expectation would be that high bb hitting teams would in general be vulnerable to low bb pitchers and this is supported by your data overall but there is a huge difference in 2015 but none in 2016, averaging out to a significant difference.
Can you think of any reason for the differences between 2015 other than random fluctuation? (obviously there is no reason to believe that general player development and maturing would specifically affect performance against low bb pitchers as opposed to other pitchers).
Chasing less pitches will help most against pitchers with good control…
Player development certainly could explain better results against one specific class of pitchers. It could explain just about anything. Before considering that though, the first question is whether there is anything to explain at all.
Good to see 15 indians fans and 4 jfg alt accounts out already.
What would be more interesting would be to correlate relative productivity with a pitcher’s control in an individual game, not in general. The issue is about dependence on walks for offensive productivity.
Any thoughts on john a’s findings, jeff? Seems kind of different from what you found.
The real question is how statistically reliable are any of these samples? Could the Cubs show severe splits against low- and high-BB pitchers one year and no split the next just by random chance alone? Does some team randomly do better against low-BB pitchers by chance? Or is the sample very reliable, such that the 2016 split indicates real, concrete improvement? Somewhere in between, meaning the best strategy is to just look at the combined sample? I think we don’t really know the error bars here.
It sounds like there was a huge split in 2015 and none at all in 2016. I guess there could be other factors involved but there are also cases of huge variation. The only way to find out is a lot more research about the relationship between high bb hitters and low bb pitchers in general, not just the cubs in 2015-6. And as i’ve been saying, what’s really needed is a study of low bb performances in individual games, based on the game, not the pitcher’s general performance. Obviouslt that would be a lot more work, but that’s the real question.
Wind blowing out. Josh Tomlin on the mound. Carlos Santana in left field?
This series could get very ugly for Cleveland very quickly.
I agree, but you have to put Santana in the field. He has to bat. Tonight will be a massive run producing night.
Why does it feel that Cleveland is in a hole and not simply “tied” with the Cubs? I don’t ask this rhetorically. I feel that that Cubs are literally up 2-1 and will simply win the remainder of the games and baseball will stop by Sunday evening. I find this perplexing.
All of the hype this series received earlier in the week: 1908 v. 1948. Blah, blah, blah. Rust belt narratives…have been pushes aside with stories about Wrigleyville’s total demise this weekend and the Cubs scoring 20 runs a game.
I wouldn’t mind this, but I also find it shortsighted and just a touch…a touch…season long confirmation bias.
well realistically Cleveland was the underdog before the series started. they have s good BP but they lost half of their rotation and were worse than the Cubs even before that.
both Boston and Toronto would have had a better chance albeit they would have been underdogs too.
I would guess it feels like that because the series is tied and game 3 chicago is a heavy favorite (at least according to most projections and vegas). I’m not sure anyone is saying Cleveland is already down 2-1 and the series will be over Sunday ….if anything the pressure is somewhat on Chicago to take advantage of this game where they are large favorites as they will be almost a coin toss for game 4.
if Cubs struggle against anyone it is against power strikeout pitchers with plus stuff and at least decent control but then again anyone is struggling against those pitchers. Tomlin throws strikes but his stuff is well below average, that is batting practice for the Cubs.
a rotation of Bauer and Tomlin behind kluber was never going to work, that rotation is one of the worst in WS history.
Tomlin is just a bad pitcher he throws meatballs and hope they don’t get clobbered too hard
Actually this rotation is about as good as the 2014 giants and they did okay.
I made one more pass at studying this question for 2016 only, but in terms of what the starting pitchers *actually did* against the Cubs, instead of their norms.
I broke all 162 starts into just two equal groups. First, by BB%:
Low-BB (4.6%) — 4.59 RA/9 … 4.1% HRs
Hi-BB (14.3%) — 6.15 RA/9 … 3.0% HRs
And then by strike %:
Hi-Strk% (66.3%) — 4.55 RA/9 … 4.1% HRs
Low-Strk% (58.7%) — 6.16 RA/9 … 3.0% HRs
Whatever else this may show (or not), the HR data are especially interesting. The better strike-throwers gave up significantly more HRs, but still came out way ahead by RA/9.
Anecdotally, now … The Cubs have so far walked much less in this postseason than the regular — down from 4.0 to 2.8 BB/game, or from 10.4% to 7.5%. But there’s still correlation between their walks and scoring:
— 3 walks or more … 7.0 R/G (6 games, 4.7 BB/G)
— 2 walks or less … 1.8 R/G (6 games, 1.0 BB/G)
And maybe that’s part of why we’re having this discussion. But the problem with reading much into that is (a) a slight majority of their walks came off relievers, and (b) the 6 starters who gave them 1 walk or less are also mainly the superstars — Kershaw twice, plus Kluber, Bumgarner, Cueto (with a battered Samardzija rounding out the list).
Tomlin’s no star. But I don’t think he’s doomed by the combo of his high-HR tendency and tonight’s wind. For one thing, he’s turned in 2 nice PS starts so far, yielding no HRs to powerful teams against whom he gave 6 jacks in 3 games this regular season. And a solo shot or two is no tragedy. More important, I think, to avoid the freebies.
The fact that low-BB pitchers gave up more homers to the Cubs is not the only relevant fact. The other fact is that Tomlin himself is extremely homer-prone. He should be expected to allow more than even the low-BB pitcher sample, just with the huge error bars associated with a single game outcome.
Also, consider this: Tomlin’s “true-talent” home run rate this game may be 1.80 HR/9, which is extremely high. However, that translates to 1 every 5 innings. If Tomlin goes 5 innings and gives up 1 home run, I think we will all be like “huh, well that was overblown.”
The other thing about high-BB giving up “fewer” homers is that a BB is generally a plate appearance much less likely to result in a homer than any other, since by definition it has the lowest ratio of strikes to balls. If a pitcher cuts down his walks, he is cutting down PA that are least likely to result in homers anyway, meaning his rate of homers per inning or homers per fly ball may stay the same, there are just more PA that may result in homers (because he is throwing strikes, some of which may be hit). The beneficial tradeoff is primarily that he is simply converting a higher % of PA into outs.
Interesting feedback, Bip. One statement I might not agree with — “a BB is generally a plate appearance much less likely to result in a homer than any other, since by definition it has the lowest ratio of strikes to balls.”
Maybe I don’t get your point, but … Doesn’t that all depend on when you snap the picture? If a pitcher eliminates a walk by grooving one in a strong batter’s count, a PA that was more likely to result in a walk suddenly becomes a PA more likely to result in a HR. (Or what am I missing?)
It’s a weird way of looking at it, I admit. On the one hand, a PA that results in a BB has a 0% chance of resulting in a homer, because we know the result and it wasn’t a homer. But, on the other hand, it is very simple. A walk is a non-homer outcome. Low-BB pitchers are pitchers who have reduced one of them, leaving the others to collectively take a higher percentage of the rest.
Basically, if pitcher A walks fewer guys than pitcher B, then when happen’s to pitcher A’s walks? Those PA just have some other results. Most likely, those PA will be distributed approximately as they already are, which means they all will increase sligtly as percentages.
Here, we might actually want to look at HR/9. If a pitcher A and B both have 1.00 HR/9 and 8.00 H/9, then pitcher A will have a higher HR%, since all their ratios are the same, but pitcher B has some extra walks also. However, I would argue that, effectively, they have the same HR rate. Throwing strikes always leads to some chance of allowing homers, but obviously getting outs is worth the occasional homer, where as a walk is *always* unproductive.
Bip — FWIW, in my study the guys with much higher HR% also had significantly more HR/9. (Left the sheet at work so I can’t give you numbers, but I did look at that before posting.)
That makes sense too. The low-BB guys probably throw more strikes in their non-BB plate appearances too, meaning they might be more vulnerable to the home run even ignoring walk rate. The main driver I would think would be survivor bias, where a pitcher who doesn’t walk a lot of guys can afford to allow more homers and still be good enough. The high-BB-high-HR pitchers don’t last.
I would think runs per game is a leaked variable here since you are basically taking the number of walks in a given game and correlating it to runs score in that same game.
I think it would be better to break the bb% in individual games into more than 2 groups. And how does the cubs’ differential compare with the league average? Thanks!!
Well, they were vulnerable to this strike-thrower on this night. Cubs draw one walk — which was really strike 3 with room to spare in every direction — and get shut out.
Tomlin was fantastic. I was shocked he came out at 58 pitches. Tito should go straight to a craps table — you can’t roll any hotter than he was tonight.
Everyone pointing out Tomlin’s homer problem was right. He only missed his expected total by one.
After being shut out 4 times in their last 8 games but scoring quite a few runs in the other 4, it’s hard not to suspect that *some* kind of pitchers give the cubs very serious problems. This is a pretty whopping standard of deviation. Control seems to be a natural parameter to test but it could be something else. I don’t remember anything like this in the postseason since the 1960 pirates.
The same sort of thing happened in the nlds. The Cubs’ hitters were stifled by Cueto (3 hits in 8 innings, one solo homer by about a foot), exploded for 4 runs in two innings against Samardzija when he decided not to throw his curve, then went to sleep for 29 innings (Ross’ solo homer was the only normal run they scored, they got 2 pitchers’ homers, Bryant’s 355-foot pop fly off the top of the wall, and a three-base error on a ground ball plus a sacrifice fly), then exploded again in the 9th inning of game 4 against Law and Romo who had pitched 2 innings the day before and Lopez who couldn’t throw a strike. This is not a normal pattern of offensive production.