Here's What a Postseason Start Looks Like in 2020

[Tom Froemming]

This is the new normal. Postseason pitching staff management has changed, as starting pitchers going deep into games becomes more and more rare. Here’s a look into a typical postseason start leading up to the World Series this season.The information below is from all rounds of the 2020 postseason leading up to, but not including the World Series so far. That’s 47 games, or 94 total starts we’re looking at. As with any starting pitching stats in 2020, the fly in the ointment is the opener. Just something to keep in mind as we take a look at some postseason starter stats.

 

Pitches Thrown

Average: 75.1

Median: 79

High: 110 Jack Flaherty, STL

 

Something odd to note is teams that had a starting pitcher eclipse 100 pitches had a losing record over these rounds, going 7-8. If we look at all games in which a starter threw 90 or more pitches, those teams went 17-15.

 

Kenta Maeda’s Game 1 start in the Wild Card round was one of those games where a starting pitcher threw 90 pitches and his team lost. José Berríos logged 75 pitches for the Twins in Game 2.

 

The most surprising thing to me is a starting pitcher threw 90 or more pitches at just a 34 percent clip in these rounds. So it’s not that 90 is the new 100, it’s more like 80. Here’s a chart that shows the number of starts made by pitches thrown.

 

Download attachment: PitchesNew.png

 

Outs Recorded

Average: 13.2

Median: 14.0

Mode: 15

High: 24 Clayton Kershaw, LAD

 

It’s not like starting pitchers are throwing fewer pitches because they’re efficient. The mode number of outs recorded over these rounds — the result that appeared most often — was 15, or five innings. That just happened to be exactly how long both Maeda and Berríos pitched in their postseason starts.

 

Bullpens are doing a lot of heavy lifting, this is becoming especially true in the postseason. The chart below shows the number of starts by outs recorded in these rounds.

 

Download attachment: OutsNew.png

 

Batters Faced

Average: 18.9

Median: 20

Mode: 19

High: 31 Max Fried, ATL

 

At this point, you shouldn’t be surprised to see a lot of decisions regarding traditional starters are related to the “third time through the order penalty.” The vast majority of starting pitchers were given enough leash to start that third trip around, but very few finish it.

 

Is this “penalty” the baseball equivalent to the boogeyman? Hitters actually had a lower OPS in their third plate appearance against a starting pitcher this year than the second time seeing them (a .765 OPS the second time, .754 the third).

 

Then again, maybe that’s just the result of a weird season. Or maybe managers being more cognisant of the dangers of allowing pitchers to go through a lineup a third time is exactly why that number has gone down. After all, in 2019 hitters had a .777 OPS the second time they faced a starter and an .807 mark the third.

 

The mode of this data set was 19 batters faced. This was the exact number of batters Berríos faced in his outing versus Houston, while Maeda faced 20 men. Anyway, here’s a chart with the number of starts broken up by batters faced.

 

Download attachment: Batters Faced New.png

 

One last item, this one breaks things up by the winning and losing team. This is a bit of an unsophisticated way to look at things, seeing as a team could pitch like trash and still win a game if their bats carry them, but there was a buy three nerdy baseball graphics get one free sale. So here you go:

 

Download attachment: WinsLosses.png

 

MORE FROM TWINS DAILY

— Latest Twins coverage from our writers

— Recent Twins discussion in our forums

— Follow Twins Daily via Twitter, Facebook or email

 

Click here to view the article

[Nine of twelve]

 

This is the new normal.
...managers being more cognizant of the dangers of allowing pitchers to go through a lineup a third time is exactly why that number has gone down. After all, in 2019 hitters had a .777 OPS the second time they faced a starter and an .807 mark the third.

And this probably should be the new normal.

I have grown tired of hearing Bert and Jack complain about "kids these days". We have phased into the era of a pitching staff functioning as a team rather than relying on today's starter. A 26-player roster with 13 or even 14 pitchers allows the manager to bring in fresh arms more frequently. Certainly there will be exceptions based on how a starter is performing on a given day but in general it makes sense to me to use several relievers who can let it fly for an inning instead of using a pitcher who has already thrown 70+ pitches when the game gets into the sixth inning or so.

It's analogous to track. The times in a 1600-meter run will always be much slower than those in a 4x400-meter relay. Starters are distance runners and relievers are sprinters. Better overall to use your sprinters when you can.

[mikelink45]

And now lets have a 30 man roster so we can keep adding more and more pitchers.  In 2019 they played with the rule that allowed players to come up and down with the 10 day IR.  That changed and rosters went up.  RP having to throw so many innings in so many games wear out.  I would prefer a 4 pitcher rule - choose four and that is all  you have for a designated game until 10 or more runs is scored.  

 

Baseball is getting boring - give me a Koufax, Gibson...BP games are not for me.  

 

 

[USAFChief]

I think the 28 man roster influenced pitching decisions a great deal.

 

It's certainly true that stater innings have been going down across baseball for years, but this year is an anomaly.

[terrydactyls]

So if your starting pitcher has retired 18 straight batters and has thrown 70 pitches, the "metrics" tell you to yank him to avoid the "third time though penalty"? What a load of #$@. Why would you replace a known quality with someone from the bullpen whom you haven't seen throw a pitch today? Analytics are taking instinct and common sense out of baseball and is trying to make a Nintendo game out of it. And I do realize that the Nintendo reference dates me. I was going to say Intellivision (which I owned) but I figured not many of you would know what that was.

[Tom Froemming]

 

I think the 28 man roster influenced pitching decisions a great deal.

It's certainly true that stater innings have been going down across baseball for years, but this year is an anomaly.

Could be, but I think the lack of off days combated that a bit. I'm a bit surprised we didn't see starters go deeper given all these back-to-back-to-etc. games on the schedule this year.

[Tom Froemming]

 

So if your starting pitcher has retired 18 straight batters and has thrown 70 pitches, the "metrics" tell you to yank him to avoid the "third time though penalty"? What a load of #$@. Why would you replace a known quality with someone from the bullpen whom you haven't seen throw a pitch today? Analytics are taking instinct and common sense out of baseball and is trying to make a Nintendo game out of it. And I do realize that the Nintendo reference dates me. I was going to say Intellivision (which I owned) but I figured not many of you would know what that was.

No, if a starting pitcher literally has a perfect game going through six innings I'm confident zero managers would remove that pitcher unless there was some kind of workload/injury concern, like the guy was pitching on super short rest or something like that.

[GNess]

Do what works best - the object is to win the game you are playing and win as many as you can during the season.  I don't mean to appear snarky, but we shouldn't over complicate this issue.

 

The team's decision-makers primarily use their experience, observation of others and available applicable data to determine how best to accomplish these two outcome based goals. 

[Hosken Bombo Disco]

This is an important question.

 

Unless my eyes are deceiving me, the Twins look like they are willing participants in the movement to downgrade the value of starting pitching. This necessarily means that relief pitching must become more important. Which starts to cast doubt on (for me, anyway) on the trade they made for starting pitching last February.

 

And yeah I would have liked to seen all the Twins starters pitch at least one more inning in their postseason starts, 2019 included.

 

/rant

[joefish]

Old schooler here. It's hard to argue against these, or most, stats.

But I can ignore them.

[ashbury]

Old schooler here. It's hard to argue against these, or most, stats.

But I can ignore them.

Well sure. But if you have a weather forecast to look at and you find it hard to argue against the prediction of rain, if you ignore it you will still wind up wet.

 

[twinfan]

Well if baseball would adopt my rule that a batter can only have 2 fouls after two strikes, you would have starters going deeper and games going faster. There are several 8-10+ pitch at bats which greatly increase the number of pitches thrown per inning and game. I also feel that, at least for this year with very few days off-especially in the post-season, you can't have a starter go very far because you will need him again soon. However, during a normal season there is no reason to have a starter throw less than 90 pitches unless they are getting hammered. Pitchers today aren't as good as pitchers in the past because they don't vary the game plan and thus can't get through a lineup the third time. And several hitters know how to spoil many 2 strike pitches to prolong the at bat.

[twinfan]

 

So if your starting pitcher has retired 18 straight batters and has thrown 70 pitches, the "metrics" tell you to yank him to avoid the "third time though penalty"? What a load of #$@. Why would you replace a known quality with someone from the bullpen whom you haven't seen throw a pitch today? Analytics are taking instinct and common sense out of baseball and is trying to make a Nintendo game out of it. And I do realize that the Nintendo reference dates me. I was going to say Intellivision (which I owned) but I figured not many of you would know what that was.

Hey. I'm an Atari man myself- not to mention a Radio Shack TRS-80

[tony&rodney]

A roster of 25 players may provide a return to more balanced decisions. There are managers who allow and pitchers that are able to successfully throw 100+ pitchers. Francona and Cleveland come to mind and we hear that Bauer wants to throw every fourth day. The game continually evolves and changes. Teams and players adapt and change to match the scenarios facing them each year. Houston swings hard until two strikes and then puts the ball in play. The Dodgers do the same. 

Nevertheless, there is a time to lift a pitcher and manager judgment and knowing the players makes a difference. In high school there is no reason to risk a pitcher's arm. I lifted a pitcher with a perfect game after 5 innings because it was early in the season and we had a rough pitch count. A kid can play baseball until his 50s. The kid thanked me and said he was gassed. Now he dominates Town Ball in his 30s. The pros are different because money plays a major roll in game decisions. Nobody wants to ruin a player's arm with overuse. So the dynamic is fluid in today's game and when MLB expands the roster the usage of pitchers is effectively changed through additional options. My own thought is that Maeda and Berrios would have been more effective than a reliever but the manager, in concert with the coaches, would be more attuned to that decision than any of us. Hopefully, the decisions are based on observations such as control and movement rather than on the third trip through the order. Certainly we all must be noting how hitters are managing to square up a fair amount of pitches from relief pitchers at near 100 mph. I'm not a fan of the three batter rule for pitchers or the expanded rosters. I would rather see a limit of twelve pitchers on the roster and a return of the fifteen day injured list. In any event, the hitters must adjust to reduce the empty at bats and futile three pitch strikeouts. A diehard baseball fan of 60+ years will never tune out but I have many conversations with younger fans turned off by the walks, strikeouts, and homeruns that result in 3-4 hour games with little action from their point of view. It is an opinion that baseball needs to consider in the 21st century.

[terrydactyls]

Well sure. But if you have a weather forecast to look at and you find it hard to argue against the prediction of rain, if you ignore it you will still wind up wet.

 

But how often are weather forecasts correct. My method is to look outside and see if the sidewalk is wet and I am usually right.

[Nine of twelve]

 

But how often are weather forecasts correct.

The majority of the time. And similarly if a baseball strategy works the majority of the time then it makes sense to employ it.

[IndianaTwin]

 

Hey. I'm an Atari man myself- not to mention a Radio Shack TRS-80

 

Pong rocks!

[IndianaTwin]

Sorry, not able to quote properly today, but this is from the original post: 

 

"Is this “penalty” the baseball equivalent to the boogeyman? Hitters actually had a lower OPS in their third plate appearance against a starting pitcher this year than the second time seeing them (a .765 OPS the second time, .754 the third)."

 

I assume that you're talking about all pitchers who pitched a second or third time through. One factor here is that the only people who got to face a batter the third time through are those that have been effective that day. It's a SSS, but my hunch is that the guys who contributed to the .754 the third time through probably were better than .765 on their second time through. 

[Nine of twelve]

 

Sorry, not able to quote properly today, but this is from the original post: 

 

"Is this “penalty” the baseball equivalent to the boogeyman? Hitters actually had a lower OPS in their third plate appearance against a starting pitcher this year than the second time seeing them (a .765 OPS the second time, .754 the third)."

 

I assume that you're talking about all pitchers who pitched a second or third time through. One factor here is that the only people who got to face a batter the third time through are those that have been effective that day. It's a SSS, but my hunch is that the guys who contributed to the .754 the third time through probably were better than .765 on their second time through. 

Yep. We would need to know the OPS against the second time through for ONLY those pitchers who pitched the third time through.

[ashbury]

One factor here is that the only people who got to face a batter the third time through are those that have been effective that day.

This is such a good observation that it deserves boldface. So many statistical conclusions we wish to draw depend on independence within the data, and when that independence is lacking then the conclusions are suspect and must be re-evaluated. The link I offer there goes off into the weeds, but perhaps the opening sentence contains enough to give the gist: "...update the probability for a hypothesis as more evidence or information becomes available".

 

[SF Twins Fan]

I really didn't want the Dodgers to win, but I was so happy the "analytical" move blew up in the Rays face.  There was absolutely zero reason to take Snell out of the game at that point.  He was cruising at 73 pitches with 9 strikeouts and only 2 hits.  The next three batters due up were 0-6 with 6 strikeouts. Mookie Betts numbers against right handers pitchers was SIGNIFICANTLY better than it was against lefties.  Additionally, Nick Anderson had a brutal postseason having given up at least 1 run in 6 straight games entering play last night.

 

Overall it was a terrible decision by Kevin Cash and one that deserves a better explanation than, "I didn't want Snell to pitch against those guys a third time."