The Future of Pitching

[True Blue Brew Crew]

Posted this in today's game thread but it probably warrants discussion. Most likely won't get much agreement so it will take lots of time to be proven true. Here goes anyway...

 

This level of bullpenning isn't sustainable for a full season

 

It sure is. It just means scrapping traditional roles. No more starters going every 5 days. No more relievers making 80-90 appearances. It's 11-12 guys going 2-3 innings at a time with typically 2 days rest each in between. There are 1458 innings plus the occasional extras to cover in a 162 game schedule. That's 121.5 innings per pitcher on a 12 man staff. An end to 200+ innings on an individual arm is a good thing.

 

The obvious retort is that you don't have 12 equal guys so it would be impossible to spread the workload evenly. What people don't realize is how many MLB pitchers carry around the same stats the first time through an order. The elite guys carry a .500-600 OPS against the 2nd and sometimes even 3rd time through. But most relievers carry similar numbers their 1st time through. There would still be a place for the elite guys to average more innings. The best of the best would push 3-4 innings. Even then, they'd still top out around 150 innings.

 

I encourage anyone who doubts this to take a look at the splits for major league pitchers. If you're good enough to stick, starter or reliever, chances are your OPS against in under .700 the 1st time through. The difficult part of the current traditional model is finding guys who hold up multiple trips through the order. There's pitching dominance to be had giving batters only 1 look.

 

It's just a matter of how long it takes for the transition to be complete. The current model of a 5 man rotation with starters called on the make 80-100 pitches every 5 days and supplemented by relievers who throw 20 pitches 3-4 times a week will be extinct in a decade. Either that or a rule change will have to be made to keep the game the way its been in the traditional sense. The change should be allowed though because it's better for the pitching arm.

 

It's nice to see the Brewers ahead of the curve on this and being one of the teams at the forefront. It might even get them a World Championship in a few short weeks.

[DuWayne Steurer]

There is no such thing as a 12 man staff. Look at any page on B-R. Most teams get through a season with 25 - 30 pitchers, or more.

 

Ostensibly, you might have 12 men on your pitching staff on any given day, but teams have really learned how to utilize the 40 man roster, options, and all the rules to shuffle players back and forth to continually have fresh arms on game day rosters.

[True Blue Brew Crew]

There is no such thing as a 12 man staff. Look at any page on B-R. Most teams get through a season with 25 - 30 pitchers, or more.

 

Ostensibly, you might have 12 men on your pitching staff on any given day, but teams have really learned how to utilize the 40 man roster, options, and all the rules to shuffle players back and forth to continually have fresh arms on game day rosters.

 

Well dipping beyond the Opening Day 12 only makes this easier. And to add to your point, this new model going forward will likely lead to fewer injuries and less turnover. Which in turn lessens the need to use 25 pitchers to get through a season. One of the reasons the Brewers used so many pitchers in 2018 was to limit the amount of looks the league got at them. There was plenty of non-performance and non-injury related shuffling. It's all about keeping MLB hitters off balance, limiting looks, and allowing pitchers to throw their best pitches. Going through an order multiple times successfully usually requires a 3rd or 4th pitch. Going through once, only requires 1-2.

[Ennder]

I would imagine the 2-4 inning outing is going to become a very common thing and the 5+ inning out is going to be much rarer in the future. The 1 IP outing will also be rarer though.

[FVBrewerFan]

Side effects are trying to figure out what a pitcher is worth in FA, fair value in a trade, and arbitration is going to be difficult.

[adambr2]

Side effects are trying to figure out what a pitcher is worth in FA, fair value in a trade, and arbitration is going to be difficult.

 

It's long overdue anyway for arbitration panels to emerge from the 1970s and start figuring out ways to pay guys based on performance measures that are actually relevant in the 21st century.

[JimH5]

I think it's sustainable from a physiology and winning perspective. But only if it is acceptable to players from a financial perspective.

 

None of the Brewers pitchers are being paid at the top of the market, and that's understandable, based on workload.

 

But few will want to come or stay here if more money can be made elsewhere.

[GAME05]

I only see the Brewers picking up on this even more next season, considering they'll have so many #5 starters, with a lot of other teams following behind. The tough part is how fickle middle relief is season to season, so it will probably be hard to see consistency on a yearly basis. But middle relievers are relatively cheap on the FA market (though getting worse considering what Colorado did), while at the same time starting pitching is so hard to find. I don't see it happening yet with big-budget teams, but I could see the mid-budget ones picking up on the trend.

 

Would pitchers really dislike it? Hader could certainly be moved to more of a two-inning closer's role, and he'd be looking at $20M per if he were a FA.

 

Though I have a hard time seeing a team do bullpen games with the entirety of their staff for reasons the Brewers are going to see at Los Angeles. Teams are playing six days in a row, or worse, the schedule the Cubs had in September. "That averages 160 innings per year" is great, but on a day-to-day basis, it'd be so easy to burn bullpen arms with a few tight games in a row playing six games per week. I think there would still be a need to have one or two traditional starters with a history of being able to go long into a game.

 

But all in all, new trends tend to pick up on where the value is in free agency and developing players, and isn't that currently found with bullpen arms?

 

Also, I bet Brooks Kieschnick is kicking himself that he showed up in the League 10 years too soon.

[True Blue Brew Crew]

Where this will really pick up steam is with the lower revenue ballclubs. If you can't afford a bunch of aces and can't come close to the payroll of the LAs, NYs, Bostons, and Chicagos of the world, your best shot is to get Ace-like numbers from guys going once through the order. And as with Moneyball and analytics, once the small market teams have success with it, the bigger markets will follow suit.

[Thurston Fluff]

I think the compensation will be there for the pitchers no matter what. Even now relievers are paid well above what they used to. Look at how much the Red Sox paid in players to get Thornburg. Does anybody think they would be less afraid to spend a lot of money for Hader in free agency? The good thing about the Brewers being ahead of the curve on this is they get these soon to be unaffordable players on the cheap until the Yankees of the world start paying top dollar for them. Then we'll be forced to move onto the next undervalued thing.

[Lathund]

The reason that the 5-man rotation has stuck around, and won't die completely any time soon, is that with limited roster space it's the most efficient way to eat innings over a 162-game season. Pretty much any other setup requires more arms to get the job done. Using the 40-man more fully can help with this. But I don't think it'll be a strategy that reduces the total number of pitchers much, if at all. Another thing that will delay this transition is that you need to find those 12 guys who can pitch in this role. Consider that most pitchers have been trained for, and used in, either the typical starter role or a more typical 1-inning reliever role. How big of an issue this'll be remains to be seen, but I don't think every reliever out there can transition as easily as we'd like.

 

Another issue I see is the workload. It's not just a matter of total innings or total pitches, but also the amount of rest inbetween. Consider that 1-inning relievers rarely go much beyond 70 innings, and extremely rarely go 80, and seem to have fatigue issues if they go further. Going 2 innings every 3 days would likely allow more innings, but I think that 150 innings when going every 3rd game would result in more injuries than 200 innings going every 5th game, so if you're using the every 3rd game schedule, you'd need to stick to the lower end of the outing lengths to get any benefit when it comes to injuries.

 

Also consider the pitching staff in the OP. 12 guys pitching every 3 days need to go 2 1/3 inning each on average to cover the full game. Whether the 13th man is an elite 1-inning guy used situationally, or a multi-inning guy to cover for days when the three designated guys get chased early. Now the problem with this comes on those days; there's no real margin of error, and you'd immediately have to send someone down and get a fresh arm.

 

Anyway, all that being said I think there are benefits to be had in terms of durability and performance in having relievers on more of a schedule rather than irregular use, and that not pitching on consecutive days can have big positive effects. The main drawback of it all though, is that having every reliever scheduled like this means you can't choose which reliever to use when. You'd end up "wasting" your best pitchers in blowouts fairly often. And this is a big, big drawback.

 

All this is not to say that I disagree with the fact that pitcher usage is changing, nor that it shouldn't. Just that there is at least some ways in which the traditional rotation actually makes sense, and that a straight up dismantling of it into a system like above has plenty of drawbacks. What will happen, and already is happening in Milwaukee, Tampa and Oakland and elsewhere to a lesser extent, is to not try to shoehorn pitchers into these moulds. If you have pitchers that fit the traditional starter usage, you should still use them that way; it's how you get the most out of them. But if you don't, then logic would dictate that you should adapt to that. Which the Brewers are doing.

 

I think that the first step would be to replace one or two spots in the rotation with some alternative setup. There aren't 150 starter-calibre pitchers in the league at any one time, so why pretend there are? What I think will become very common is to eliminate the 5th starter. He already gets skipped around off days by many teams, and it's not a huge step to instead bullpen or use some piggyback type of situations on the days where a 5th starter is needed. So now you only really need to find 4 starters. Some teams will go further of course, but I think that it's worth keeping in mind that if you have starters who can go deep into games and perform well, those outings will only make your other outings stronger. You'll better be able to control reliever availability, have them less fatigued when you need them the most and such. I'd imagine starts getting shorter overall, but the Scherzers of the world will still keep doing pretty much the same thing they're doing. Teams just won't have mediocre starters trying to replicate that, but instead adapt. And multi-inning relievers will most definitely become more common again.

[True Blue Brew Crew]

Josh Hader is showing the folly in having your best arms impact only 1 of every 5 games. You say there will always be room for a Scherzer every 5 days. I say there's even more use for a Scherzer locking down a game every 2-3 days. Not in the sense of being used as a 1 inning closer. But to be used as a guy you introduce into a game you have the lead to shut down the heart of a lineup and then stay on through to the last out. Today's aces will be future guys who come into games your team has the lead and then cover the final 3-4 innings. Instead of 30-35 starts, they can impact 60 games.

[True Blue Brew Crew]

Here are the current Brewers' pitchers 2018 opponent splits the 1st time through the batting order...

 

• Pitcher - AB AVG OBP SLG OPS
   
  Burnes - 135 .200 .273 .326 .599
   
  Cedeno - 124 .210 .300 .290 .590
   
  Chacin - 297 .192 .260 .300 .560
   
  Gonzalez - 260 .246 ..332 .335 .667
   
  Guerra - 228 .225 .303 .352 .656
   
  Hader - 270 .130 .218 .252 .470
   
  Jeffress - 269 .182 .258 .271 .530
   
  Knebel - 196 .194 .287 .372 .659
   
  Miley - 129 .225 .286 .310 .596
   
  Peralta - 119 .176 .315 .294 .609
   
  Soria - 236 .225 .280 .339 .619
   
  Woodruff - 126 .206 .265 .317 .582

 

Highest OPS against is .667. Not a single OPS above .700. Some teams now see that they can absolutely shut down offenses by not giving batting orders a 2nd look at a pitcher on a given day. This is no doubt the future.

[True Blue Brew Crew]

Look at those batting averages and on base percentages allowed! Especially if you took out Gonzalez's numbers most of which were accumulated with Washington. Batting averages in the low 200s and on base percentages below .300. The first team to successfully transition their pitchers to this form of usage is going to lead MLB in every pitching category. And it won't require high priced arms.

[coolhandluke121]

Posted this in today's game thread but it probably warrants discussion. Most likely won't get much agreement so it will take lots of time to be proven true. Here goes anyway...

 

This level of bullpenning isn't sustainable for a full season

 

It sure is. It just means scrapping traditional roles. No more starters going every 5 days. No more relievers making 80-90 appearances. It's 11-12 guys going 2-3 innings at a time with typically 2 days rest each in between. There are 1458 innings plus the occasional extras to cover in a 162 game schedule. That's 121.5 innings per pitcher on a 12 man staff. An end to 200+ innings on an individual arm is a good thing.

 

 

I've been on this train since this time last year. I mistakenly called it "piggybacking" which unfortunately caused a lot of people to dismiss my ideas completely. Strict, scheduled piggybacks are now what I was advocating. I merely wanted to blur the lines between starters and long relievers more, and I provided evidence for why it would work. Starters were working too many times through the order and relievers, who are often just failed starters, often have better stats despite ostensibly not being as good. This shows that the balance of workload between the two groups is not in the sweet spot. It also shows that guys don't really have to know exactly when they're going to pitch, because relievers never really know when they arrive at the park if they're going to work. And lest we forget, you often waste a plate appearance by letting a pitcher hit for the 2nd or 3rd time. I think guys could routinely start on 3 days rest if they're capped at 60-70 pitches, with more long relief. I think relievers could give you more innings total if you ask for multiple innings spread out over fewer appearances. And of course you can use shuttle guys to make it work for staffing.

 

I do think more of this is sustainable when you have enough guys with options, which the Brewers do. I truly believe this is the start of a revolution and I'm thrilled that the Brewers are spearheading it. I said at the beginning of the year that there would be a lot of doubters, but that it would be adopted as common practice once everyone else saw how well it worked - and I had faith that it would.

[True Blue Brew Crew]

When I get a chance, I'll post the numbers for the 2nd time through and 3rd time through a lineup. Hint: they're often ugly. There are many reason for this. Fatigue, familiarity for the batter, and most importantly - the need for the pitcher to rely more on their 3rd and 4th best pitches. Taking away the need for multiple trips through the order means much less need for a often weak 3rd or 4th pitch. And in turn much more use and opportunity for pitchers who just never are able to develop a 3rd or 4th pitch.

[True Blue Brew Crew]

Let's take a look at Ben Sheets, the closest thing to a homegrown ACE the Brewers have had this century. Here are his career numbers for the 1st, 2nd, and 3rd time through a batting order...

 

• 1st BA OBP SLG OPS
   
  .241 .283 .398 .681
   
  2nd BA OBP SLG OPS
   
  .260 .306 .420 .725
   
  3rd BA OBP SLG OPS
   
  .272 .317 .449 .766

 

EVERY SINGLE CURRENT BREWER PITCHER has been better their 1st time through an order this year than Ben Sheets career 2nd time through the order.

[Lathund]

Josh Hader is showing the folly in having your best arms impact only 1 of every 5 games. You say there will always be room for a Scherzer every 5 days. I say there's even more use for a Scherzer locking down a game every 2-3 days. Not in the sense of being used as a 1 inning closer. But to be used as a guy you introduce into a game you have the lead to shut down the heart of a lineup and then stay on through to the last out. Today's aces will be future guys who come into games your team has the lead and then cover the final 3-4 innings. Instead of 30-35 starts, they can impact 60 games.

 

You need both guys who come in to lock down games, and guys who get you to that point in the first place. Max Scherzer can do almost what Hader can do, except in 220 innings rather than 80. He should stay in that role. Hader got better as a reliever, because he could get by using only his two good pitches, and because he could go all out for shorter bursts. Hader is exactly the type of pitcher who should be a reliever rather than persist as a starter, because they're simply better in that role. But the guys who actually can reliably perform to that insanely high level over 200+ innings should keep doing that. That type of talent is in short supply, and whoever you get to pitch those remaining 120 innings will do a worse job of it.

 

Times through the order penalty is real, we all know (or should know) that, it's not exactly news. But when you have someone who performs the same the third time through the order as he does the second (Like Scherzer has for his career), then why would you cut his innings total in half? Having one guy like that means you don't have to use your 2-3 inning guys that day, and you can use them better on the 4 days where you don't have Scherzer.

[True Blue Brew Crew]

Yes you absolutely would cut their innings. You'd cut them so they can impact twice as many games. They don't award wins on the field based on innings pitched. And the bonus would be fewer arm injuries. The best of the best, like Scherzer would still lead the team in innings. He'd also lead the team in appearances. Both would be less taxing on his arm than the current usage.

[True Blue Brew Crew]

These numbers also show that many, many more pitchers would be classifiable as "dominant". If the 50-80 fewer innings Scherzer pitches are covered by guys getting the same results, you're not sacrificing anything. You're simply using Scherzer in the spots of most leverage, impact, and value. In other words, exactly the way the Brewers used Hader this year.

[coolhandluke121]

Yes you absolutely would cut their innings. You'd cut them so they can impact twice as many games. They don't award wins on the field based on innings pitched. And the bonus would be fewer arm injuries. The best of the best, like Scherzer would still lead the team in innings. He'd also lead the team in appearances. Both would be less taxing on his arm than the current usage.

 

I'm not sure I'd go that for. There's a cost of warming up and getting ready to appear in a game. Once you're out there and you're cruising, it's not hard to go 7 or 8 innings. The downside is not arm injuries, which happen to relievers too. It's just that most guys can't do it and be effective. You could probably get a few more games out of Scherzer, but I think he's at the point where you'd have to accept less total innings if you did that. It's not like you have 200+ innings to play with there. Clearly the best relievers can't pitch 100 games with 2 innings each game. It's not a simple linear relationship between appearances and innings per appearance.

[clancyphile]

These numbers also show that many, many more pitchers would be classifiable as "dominant". If the 50-80 fewer innings Scherzer pitches are covered by guys getting the same results, you're not sacrificing anything. You're simply using Scherzer in the spots of most leverage, impact, and value. In other words, exactly the way the Brewers used Hader this year.

 

Hader is different from Scherzer or Kershaw - he only has two pitches he can command.

 

If you can get someone with three or four pitches, and that kind of talent, go for the TOR option. Ideally, develop both shut-down relievers who can go two innings AND good starters.

[Lathund]

The system you propose has pitchers essentially pitching on a schedule, just like a normal rotation only every 3rd day, whether you realize it or not; it's the only way you can get it to work over a long stretch of time with all the innings you have to cover, you need all 12-13 pitchers over every 3-game cycle. There is very little flexibility in when to deploy which pitcher. Which takes away the main advantage you have with a Hader; that you can choose when to deploy him, i.e when he can improve your chances of winning the most.

 

I'm not arguing against the concept of limiting multiple times through the order, or more use of multi-inning relievers (That's one thing I've championed for a long time). What I'm saying is that the practical realities of a 162-game season is that you can't stick to this system every day with the current roster size and with how current pitchers are conditioned. The pitchers who can do this 2-3 inning role for the most part won't be the pitchers who are relievers today, but starters. And there aren't enough of them. Relievers who have a sub-700 OPS against when facing 3-4 batters per appearance and throwing 15-20 pitches (i.e what typical relievers like Knebel and Soria have done to produce the numbers quoted above) likely won't do the same against 8-9 batters and 40+ pitches.

 

If the 50-80 fewer innings Scherzer pitches are covered by guys getting the same results, you're not sacrificing anything

 

You're not gonna pitch 140-170 innings pitching every 3 days. That won't reduce injuries, that'll end his career. No, what you're losing is 100-120 innings. And you wouldn't sacrifice anything? Seriously? Scherzer this year had a 2.53 ERA. The Nationals bullpen had a 4.05 ERA, which would have come mostly from 1-inning appearances. You're saying that if you took that bullpen, had them pitch ~50% more innings, and in longer stints, that that would drop their ERA to 2.53? Or, to be generous, to maybe 2.75 or so as Scherzer would likely drop his ERA too?

 

When you have a pitcher like that, who over that many innings already gives up runs at a lower rate than like 90% of all relievers do, it makes absolutely no sense to have him pitch less. Not unless you had a way to make sure he pitched those fewer innings only when they mattered the most. But to do that, to get that freedom to be able to deploy him when you want, and not when you need to, you can't have a situation where he'll be needed to pitch on a schedule, or close to it; if he's on a schedule anyway, you might as well get 220 innings instead of 100. So to put him in that position, you need to have some of the other relievers pitch more innings, in order to preserve your elite pitcher some. And then you're in a situation where you're no longer giving the most innings to your best pitcher. And that's fine up to a point, but you can only get so far with that. Reducing the total numbers of runs you give up is what improves your chances of winning the most games over a full season the most. Which is why run differential or pythagorean record are better at predicint the rest of the season than the current win totals are. Shifting around where and when you give up those runs can and will help you, but only to a certain point.

 

Basically you need bulk innings from somwhere, in order to make the most out of your Haders. The more of those innings that come from lesser pitchers, the more runs you'll give up and the more losses you'll get. Traditional pitcher usage has plenty of flaws, but the general principle of trying to have good pitchers pitch as many of your innings as possible is a good one. In a world where Josh Haders grew on trees and you could get 13 of him, you'd probably be best off having them pitch 2-3 innings every 3-4 days. you're right about that. But you can't fill a team with guys like that. There will be a wide range of talent on a team, and you need to deploy that talent in whatever way gets you the most wins. Dividing the playing time equally and randomly will not achieve that.

[True Blue Brew Crew]

When literally every capable major league pitcher carries an OPS against under .700 their first time through a lineup, right there that's your bulk. Let the Ben Sheets numbers sink in. Literally every Brewers pitcher right now is better coming in fresh than Ben Sheets going a 2nd time through. This isn't a proposal or a hypothetical. It's already begun. The only question is how long until it becomes the new norm.

[True Blue Brew Crew]

Even for the rare pitchers (true Aces) whose numbers hold up 2nd and 3rd times through a batting order, they become more useful when you can use them in high leverage roles 2-3 times every 5 games as opposed to just 1 every 5. When a middle reliever limited to one trip through the order posts the same sub .600 OPS that Scherzer does, you can now find more useful ways to have a Scherzer impact more games.

 

Now obviously the real Max Scherzer is not likely to transition from an Ace to a mere out getter. This is going to take time to implement. And the last places to embrace this will be the big markets. Big ticket #1s will continue to migrate to the biggest markets. But as small markets embrace the effective subdivided use of their entire 12 man staffs, and benefit from the result, the idea of forcing yourself into a 5-man rotation when you're lucky if you have even 1 guy capable of going deep into a game effectively, is foolish pursuit. As the number of pitchers called upon to pitch 5-7 innings dwindles, pitchers will be groomed for 2-3 inning, more frequent appearances. Statistically, more than 80% of MLB starting pitchers have no business being counted on to go deep into games. So the question is, why make them?

 

It will take time to transition, but it's a foregone conclusion when the current model has most starters allowing MVP-like batting numbers when they face a lineup twice. Especially when every arm on the roster is statistically more likely to post a better result coming in fresh.