Building a Pitching Pipeline

[Mike Sixel]

4 minutes ago, Major League Ready said:

I would put virtually no weight given this small sample size.  However, I would be interested in what percentage of games started were won by the players team over the previous 2-3 years.  Tells me something about consistency and if a player is able to compete when they don't have their best stuff.  Of course, I would still want all the other stats.

I want DeGrom regardless of the individual win stat in a team game.....I think that's the point here. But, man, it would be great if that wasn't what this thread was about....

 

[chpettit19]

3 minutes ago, Major League Ready said:

I would put virtually no weight given this small sample size.  However, I would be interested in what percentage of games started were won by the players team over the previous 2-3 years.  Tells me something about consistency and if a player is able to compete when they don't have their best stuff.

What does percentage of games started by that player being won by their team tell you? I go out there and pitch to 6.58 ERA over 20 starts, but my bullpen has an ERA of 0 and my team averages 10 runs a game when I start and they win 18 of my 20 starts. What does that tell you about me? That's the point of this. Individual W-L record tells you nothing about the individual beyond whether or not a starter completed 5 innings or a reliever completed 1. Doesn't tell you if they gave up runs. Doesn't tell you if they pitched well. In fact you can give up a ton of runs and still earn a W which would put a positive in your scouting report, but would be wrong. Individual pitcher W-L records don't tell you anything at all. Does that mean people can't like seeing them or be interested in them? No. But in terms of gathering information about the skills of an individual player it tells you nothing. 

[chpettit19]

9 minutes ago, beckmt said:

As you know, this is impossible.  What else is needed, for the lower ERA, Whip was it a mirage of ball just hit at people and unsustainable, how many strikeouts did each have,  how many of his runners left on scored (bullpen issues), hard hit or exit velocity average.  The win-loss means nothing.  Also some of it is the strength of the club they are playing for.

This is the point. W-L record tells you nothing about an individual player. That's the point I've been trying to make. Are ERA, WHIP, WAR, or any other stat telling you 100% of the story on that player? No. But they at least tell you something about them specifically. Pitcher W-L tells you nothing outside of a starter going 5 innings or a reliever finishing 1.

[Major League Ready]

Just now, chpettit19 said:

This is the point. W-L record tells you nothing about an individual player. That's the point I've been trying to make. Are ERA, WHIP, WAR, or any other stat telling you 100% of the story on that player? No. But they at least tell you something about them specifically. Pitcher W-L tells you nothing outside of a starter going 5 innings or a reliever finishing 1.

It tells you very little by itself but that's now how analysis is done.  You want to insist on this point so you are conviently ignoring all else.

[chpettit19]

5 minutes ago, Major League Ready said:

It tells you very little by itself but that's now how analysis is done.  You want to insist on this point so you are conviently ignoring all else.

Where does it rank on the list of possible pitching stats and when should a GM start caring about it? Is that better? When Jacob deGrom hits free agency are you telling him he's worth Pineda money cuz he just doesn't win games? I'm still yet to see anyone on here tell me what the pitcher W-L stat tells them about that individual pitcher and makes it a useful stat. A few people are adamant I'm wrong and it's important, but none of them have told me why it's important beyond saying it's interesting when compared to other stats.

There are parameters set that dictate what goes into a stat and how it is earned. That is what the stat tells you. Pitcher W-L tells you a starter made it 5 innings, his team scored more runs than he gave up, and his pen never blew that lead. Or it tells you a reliever finished an inning in which his team took the lead and never gave it back. That is not useful in determining anything about those pitchers. If you can tell me what is useful about knowing Yusmeiro Petit has 6 wins this year (on it's own, not by adding in other stats) I'll listen and I'm more than willing to change my mind. If not then just accept it's interesting in relation to the past and to see deGrom never get Ws, but in and of itself the stat tells you nothing useful.

[h2oface]

1 hour ago, chpettit19 said:

Pitcher A- 9 GS, 2-2 W-L, 1.58 ERA, 0.74 WHIP

Pitcher B- 9 GS, 5-2 W-L, 4.69 ERA, 1.38 WHIP

Pitcher C- 9 GS, 3-5 W-L, 4.53 ERA, 1.40 WHIP

Pitcher D- 9 GS, 4-2 W-L, 2.53 ERA, 1.03 WHIP

Pitcher E- 6 GS, 3-2 W-L, 0.68 ERA, 0.60 WHIP

Pitcher F- 10 GS, 5-2 W-L, 4.39 ERA, 1.19 WHIP

Pitcher G- 9 GS, 5-3 W-L, 4.79 ERA, 1.43 WHIP

Pitcher H- 9 GS, 4-4 W-L, 2.47 ERA, 0.86 WHIP

Rate these pitchers in order of who you'd want and tell me how much weight you put into their W-L record. You're the GM of the Twins trying to win a WS and these are the guys you have available to sign. Tell me how useful their W-L records are when deciding who to sign.

To give you an answer, since you are requesting one, and you took the time to make that awesome table of pitchers, I want pitcher E, becasue it is Jacob deGrom. But if I was the Twins FO, I would select pitchers B,C,F, or G.... because that is what they do to supply new pitchers to our team. Or worse stat pitchers. I don't mind the sss for your purposes. It really doesn't matter for your mission.

Enjoy your arguing. It doesn't matter what anyone else chimes in with, you are considering it 'noise'. You hear what you want to hear. No one has said that W/L is the most important stat. Nobody, You can win if you like. I am fine with considering W/L stats not useless for my baseball considerations. You can chose what you like.

[chpettit19]

14 hours ago, h2oface said:

To give you an answer, since you are requesting one, and you took the time to make that awesome table of pitchers, I want pitcher E, becasue it is Jacob deGrom. But if I was the Twins FO, I would select pitchers B,C,F, or G.... because that is what they do to supply new pitchers to our team. Or worse stat pitchers. I don't mind the sss for your purposes. It really doesn't matter for your mission.

Enjoy your arguing. It doesn't matter what anyone else chimes in with, you are considering it 'noise'. You hear what you want to hear. No one has said that W/L is the most important stat. Nobody, You can win if you like. I am fine with considering W/L stats not useless for my baseball considerations. You can chose what you like.

The problem is neither you nor anyone else arguing for pitcher W-L has been able to tell me why it's important or useful. If your entire argument for anything in life is "we used to use it" then it's just noise. Sorry. I have no problem with you considering W-L stats useful. You can't tell me why they're useful so it doesn't make sense to me why you'd consider them useful, but that's your choice.

But this conversation got started with you saying Seth's credibility "takes a hit" because he doesn't consider it a useful stat. I explained in multiple ways why it's not useful and your response is to mock the front office while continuing to not explain why anyone should find you credible in the least when it comes to stats. If you're going to publicly question someone's credibility it may be prudent to have a better reason than "it's how it used to be." But thank you for letting me "win." Very kind of you.

[stringer bell]

Pitching has continued to evolve in the last decade. Velocity is up everywhere. Once upon a time, a team might have one guy who threw fastballs at over 95 mph and now most teams have three or more such guys. It has become understood and expected that being assigned to the bullpen will add significantly to the fastball. On the downside, injuries seem to be increasing, most likely because of the maximum effort required to throw so hard. 

In addition to figuring out which pitchers have the stuff to thrive in the majors, the front office also has to figure which pitchers are most likely destined for the bullpen, whether that is because their stuff will then "play up" or because their chances of staying healthy will increase. 

I would like to see what the "arm farm" has to offer. The long-term success of the Twins' franchise depends on homegrown talent and right now most of their pitching staff is either major league or minor league free-agent signings.

 

[Dodecahedron]

17 hours ago, chpettit19 said:

When analyzing data, the more USEFUL metrics you have the better. Some things are absolutely noise. The point of individual stats is to tell you useful information about that individual. When that metric measures far more about your team than it does about you it is noise and not helpful.

Isn't the pitcher a part of the team?  This is why you analyze further.  How much of the measurement was due to this particular individual on the team?

You want this to be a black and white issue.  It's just not.

[chpettit19]

5 minutes ago, Dodecahedron said:

Isn't the pitcher a part of the team?  This is why you analyze further.  How much of the measurement was due to this particular individual on the team?

You want this to be a black and white issue.  It's just not.

That's why W-L is a useful team stat, not individual stat. Jorge Alcala "earned" a win last night. How useful is that to telling you how he performed last night? At best it's useless, at worst it actually tells you a false story of success and you're giving him positive marks for having blown a hold situation and given the other team the lead.

W-L Records from 2019

18-4

17-6

15-4

15-7

14-12

What do those stats tell you about any of those pitchers? That's the question. When you're talking about an individual stat being useful you have to look at that stat on it's own and see what it's telling you. Who pitched well in that group and who didn't? An individual stat is supposed to tell you about that individual. None of them tell you 100% about that individual, but W-L tells you basically 0%. It gives you no meaningful information. If you still find it interesting to see W-L records that's fine. But the stat tells you nothing of use about an individual player's performance.

[Major League Ready]

17 hours ago, Mike Sixel said:

I want DeGrom regardless of the individual win stat in a team game.....I think that's the point here. But, man, it would be great if that wasn't what this thread was about....

 

You might not have noticed I modified the scenario to what percentage of games were won by the team over a large sample size so I am advocating using win/loss in a different context and with consideration to a myriad of other stats.  Individual Win / Loss has limited value but it's not a completely useless stat if properly positioned and utilized in concert with other stats.  I am not sure where you were going with the comment regarding what the thread is about.  This tangent has little to do with building a pitching pipeline.  My observation is that many here focus on block buster trades and huge free agent signings.  However, when you look at playoff teams and specifically mid market or lower revenue teams, pitching and the team construction in general are products of savvy trades for guys who have not yet been established.  In some cases that profile is players that have minimal MLB experience and sometimes its trading for Tatis Jr when he is 19.  It’s the Whitesox trading away sale and Eaton or the Royals trading Grienke.

The same kind of focus on the highest profile players extends to drafting both here and internationally.  Some here assume it’s essential to take pitchers in the 1^st and 2^nd rounds.  Where is the supporting data?  Degrom was a 9^th rounder and there countless examples like him.  The same type of assumption exists with international signees.  Several people have made the same assumption that the best strategy is signing the huge bonus guys.  Where is the data to support this theory?  I have looked back at several international drafts and the #1.5M guys have actually become great players more often than the $3M guys.  Of course, this is a gross generalization but the numbers suggest signing two or three well regarded prospects instead of the highest profile ($2.5-$4M) international prospects is the best strategy.

I thought did a nice job of profiling Cleveland’s approach and success.  We are going to find out over the next couple of years if the new regime has found and developed picks in later rounds or acquired them by trading rentals that are more than back of the rotation guys. 

[Mike Sixel]

You looked at international drafts, but not the data about the likelihood of early draft picks being good compared to late picks? 

[Dman]

21 hours ago, MMMordabito said:

Blayne Enlow - A+
Charlie Barnes - AAA
Bryan Sammons - AA
Calvin Faucher - AA
Bailey Ober - AAA
Derek Molina - A+

Those are the pitchers left from the 2017 draft (Falvine's first).  Running out of time and horses for that Bieber/Civale/Plesac emergence from that group.  Twins have four innings from Ober so far.

Meanwhile, Cleveland has just a many pitchers left from that year. Karinchak has already arrived despite the Garlick homer.  Kyle Nelson just got called up as well. 

I don't disagree with you.  I never liked the Leach pick and a lot experts thought Rooker was a reach where he was taken as well especially given his age.  Have to remember though the two biggest upside guys they got were teenagers in Lewis and Enlow so was likely going to take longer for them make it.  Lewis would probably already be playing if not for a tough year in High A ball, Covid and then injury.  looks like 2023 before a Major contribution for him.

Barnes and Sammons are your typical 5th starter lefties that will be hard pressed to make it.  Ober has a chance but unlikely to be more than a 5th starter if his stuff does actually play so none of those guys look like a Cleveland type starter to me.  It is pretty much Enlow or bust for 2017.

2018 looks a bit more promising with Winder and Sands but will have to wait and see.