Max Scherzer and the Quality of 20 Strikeout Opponents

Recording 20 strikeouts is an amazing feat — and a rare one, too: this past Wednesday, Max Scherzer joined Roger Clemens, Kerry Wood, and Randy Johnson as the only players ever to accomplish it in nine innings. What they’ve done is exceptional and, naturally, separates them from other pitchers. And while similar, each performance occurred in the context of different variables which made it distinct: era, opponent, ballpark, etc. No one is required to decide which pitcher faced the most challenging set of circumstances among those five games (Clemens did it twice). Yet, I’ve elected to make an attempt, anyway.

When first trying to understand which game posed the most difficultly, we immediately turn to the opponent. Roger Clemens, for example, achieved both of his 20-strikeout games with the designated hitter in play. Striking out 20 players of major-league caliber — regardless of how low they are on that particular scale — is still incredibly difficult. It’s more difficult to record 20 strikeouts against good hitters. The Detroit Tigers lineup Scherzer faced, meanwhile, is a top-heavy one. But solid overall, even if it didn’t feature a DH.

The table below depicts the lineup the Tigers deployed against Scherzer as well as each respective hitter’s rest-of-season projected numbers for wRC+ and strikeout percentage. (Jordan Zimmermann‘s career numbers were used, and his strikeout percentage and wRC+ were combined proportionally with pinch-hitter Jarrod Saltalamacchia’s. Weighted average accounts for the top of the order getting more plate appearances than the bottom)

Tigers Lineup Against Max Scherzer

Player	wRC+
Ian Kinsler	101
J.D. Martinez	124
Miguel Cabrera	145
Victor Martinez	119
Justin Upton	118
James McCann	71
Anthony Gose	80
Jose Iglesias	84
Jordan Zimmermann	33
AVERAGE	97
Weighted AVG	100

*Zimmerman’s wRC+ averaged proportionally with pinch-hitter Jarrod Saltalamacchia.

We see a very solid group here — a lineup that, excluding the pitcher, is above-average overall. That’s very impressive. As to how it compares with other 20-K games, the table below shows the players’ wRC+ marks during the season in which the game occurred — unless they were very low, that is, in which case career numbers were used, or an average of the seasonal numbers with career numbers depending on the disparity between the two:

Despite having to compete against the designated hitter, Roger Clemens faced a well below-average lineup in 1996 and an average one in his 1986 20-K game. The Cincinnati Reds put out an atrocious lineup in their game against Randy Johnson and somehow took the game to extra innings. That 1998 Astros team had a really good lineup, making Kerry Wood’s feat even more impressive, if that’s possible. Also, of note: Brad Ausmus, current Tigers manager, was the catcher in two of the games above, which means he has now witnessed history, and been on the wrong side of it, three times.

In terms of opponents, Max Scherzer fits along nicely with Clemens’ 1986 game, but behind Kerry Wood’s start. Of course, what the above information reveals relates to the opponent’s offense generally. These performances deal specifically in strikeouts. The tables below shows the projected strikeout rate for the players starting against Scherzer.

Projected K% for Starters in Scherzer’s 20-K Game

Player	K%
Ian Kinsler	12.5
J.D. Martinez	26.0
Miguel Cabrera	17.0
Victor Martinez	9.8
Justin Upton	26.5
James McCann	21.9
Anthony Gose	27.8
Jose Iglesias	12.2
Jordan Zimmermann	28.5
AVERAGE	20.2
wAVG	20.0

* Zimmerman’s wRC+ averaged proportionally with pinch-hitter Jarrod Saltalamacchia.

Strikeout rates over the last decade have generally been increasing, so it’s possible that Clemens, Johnson, and Wood faced more difficult lineups when it comes to striking out. While possible, however, it’s not actually the case:

We see again that Wood’s performance was truly exceptional, but the difference in eras between the players did not result in vastly different strikeout percentages.

One factor we have yet to discuss is the number of batters faced. Scherzer, facing 33 batters, had more strikeout opportunities than Wood and Johnson (29 each) or Clemens (30 and 32). We could just look at the strikeout rate and call it a day, or we could combine the information above with the total batters faced to come up with a comparable look.

First, let’s look at the rarity of 20 strikeouts, even for a pitcher like Max Scherzer. We know that Max Scherzer generally strikes out batters at a rate around 30%, which is very good. We also know that Max Scherzer faced 33 batters against the Tigers. With that information, we can calculate the percentage chance of striking out 20 batters assuming every player is likely to strike out 30% of the time. The graph below shows the results.

Max Scherzer is a great pitcher with swing-and-miss stuff, so we see his strikeouts peak in the 9-11 range. So, for a very good pitcher like Max Scherzer, even when he gets to face 33 batters (which he often doesn’t), we would expect a 20-strikeout game just 0.02% of the time — or one out of around 5,000 starts. An 18-K game is 10 times more likely than a 20-K game, and a 21-K game is four times less likely than a 20-K game. This is rarefied air.

Of course, that doesn’t really allow us to compare to other pitcher’s games because we are using Max Scherzer’s expected strikeouts. What we can do is put in numbers for the average pitcher against the lineups that each pitcher faced, then plug in the number of batters, and attempt to find how rare each game would be from an average pitcher.

Not to shock you, but these numbers are going to be extremely high. Consider the chances of a perfect game. Assuming an average team with an OBP of .315 up and down the lineup, you would need something with a 68.5% chance of happening to happen 27 straight times. The chances of this are around 1 in 27,000. With the current number of teams, that comes to around one in five years or so, and there have been 14 in the last 50 years. But when talking about strikeouts there is one advantage: they don’t have to be consecutive. But one huge disadvantage, too: hitters don’t strike out roughly 80% of the time. So accumulating large numbers of strikeouts can be incredibly difficult.

Let’s assume that every player in the lineup should normally strike out at their team’s overall weighted rate from above (which admittedly is quite the assumption), and use the number of batters faced in their game to find the odds of an average pitcher striking out 20 under those circumstances. There are two odds columns below. The first has the odds of getting exactly 20 strikeouts, no more, no less. The second one has the odds of getting at least 20 strikeouts, which includes the expected number of starts with 21 or more strikeouts. Those are rare scenarios, but they do help the odds by just a little bit.

The numbers between Scherzer, Johnson, and Wood look far apart, but in reality their significance is rather similar, as they simply represent the inverse of an unsightly number of zeros to the right of the decimal point. We see the 1-in-3-million for Scherzer and find it astonishing. If we lowered the bar to 17 strikeouts, the odds fall to 1 in around 23,000. The leap required for those last few strikeouts is huge. That Kerry Wood game is one of the best-pitched games of all time. All of these game are. An average pitcher might find himself having a good game and pitch his way to a no-hitter or even a perfect game, but under the present circumstances, it seems nearly impossible for someone to hit 20 strikeouts in a nine-inning game. You really do have to be great, like Max Scherzer.