The Crane Pool Forum

Cahna got drilled more than anybody and he neither dug in close to the plate nor leaned in during his swing.

It's hard to explain. Maybe the prevailing scouting report on the team said they were all vulnerable inside.

Good point about the scouting reports.

One thing I found interesting is the booth said one night that while Nimmo has decent speed, he doesn’t steal many bases. The thought is that he worries about getting hurt.

Marshmallowmilkshake wrote: ↑Sun Dec 11, 2022 5:56 pm Good point about the scouting reports.

One thing I found interesting is the booth said one night that while Nimmo has decent speed, he doesn’t steal many bases. The thought is that he worries about getting hurt.

Or that stealing bases is of marginal use? I forget the exact stats, but I read somewhere that if you steal at 75%, 30 bases in 40 tries, you're accomplishing very little positive for your team's offense, on the order of a run or two per season.

I would disagree with that.

I think we've all read some variation on that, and allowed it to become an established truth, while we neglect to cite where we got it, and mostly don't remember where we got it, or what the break-even point actually is alleged to be.

Edgy MD wrote: ↑Sun Dec 11, 2022 6:21 pm I would disagree with that.

I think we've all read some variation on that, and allowed it to become an established truth, while we neglect to cite where we got it, and mostly don't remember where we got it, or what the break-even point actually is alleged to be.

I second that...obviously, if a guy is not an elite base stealer, you might be better off with them not risking it...however, if you have someone like Jose Reyes, Lance Johnson, Mookie Wilson, Lenny Dykstra, Trea Turner, Rickie Henderson, Tim Raines, etc type of elite base stealing; it's another story...

getting to first is more important than not getting to first...however, getting into scoring position is more important than getting to first base...There's gotta be a margin where speed outweighs OBP, and conversely a margin whereas OBP is more important...

cherry-picking a little bit but...

Reyes averaged 66 stolen bases per 162 games from 2005-2008; his "162 game average" for those seasons results in:
116 runs scored; 68 RBI; 80 K's;

Nimmo in his career so far has an almost identical Slugging Percentage while having a significantly higher OBP (and more strikeouts as well); his 162 game average:
89 runs scored; 57 RBI; 141 K's;

yes, Reyes played on better offensive teams than Nimmo so his teammates are going to affect things; but still Reyes had a significantly higher runs per game and significantly less strikeouts...a guy like Reyes (or Rickey) in their prime, force pitchers and defense to make mistakes which IMO helps their teammates have better hitting opportunities...

A guy like Eric Young Jr doesn't get on base or hit enough XBH's for his speed to be real useful...

and a guy like Jay Payton that might have good speed but sucks at stealing bases, is better off staying put (Jay stole 14 bases as a Met, while getting caught 17 times)...by the way, I didn't realize how bad Payton was in terms of stolen bases or OBP until I just looked him up

I believe the modern game undervalues the benefits of an aggressive and capable base stealer. A base good base stealer can disrupt the defense as well as impact the pitches his teammates often see when batting behind him.

Remember the chaos Terrence Gore wreaked in his limited time with the Mets at the end of last season?

Edgy MD wrote: ↑Sun Dec 11, 2022 6:21 pm I would disagree with that.

Which part? You'd almost have to agree that there is a point where your caught stealings are costing the team more runs than your stolen bases are creating.

I think the conventional wisdom was that the breakeven point was about 67%, and I believe that figure has only risen in recent years. But let's say it's 67%--it certainly isn't far from that figure.

So someone who steals at a 75% rate, 30/40, is breaking even with 30 attempts (where he steals 20 bases) and netting you 10 bases on his other 10 attempts. How many runs is 10 bases? You know it's not 10 runs, not even close to that. Is it 5 runs? Doubt that very much. 3? 2? Something like that.

So how many wins is your 30/40 stealer getting for you, if his stealing attempts add up to 3 runs or fewer?

If you'd like to disagree with my speculations here, could you start by identifying the single largest flaw in my reasoning? Do you think the breakeven point is much lower than 67%? Do you think 10 bases over the course of 162 games are much more productive than I'm making them out to be?

I don't know what's to agree with or disagree about. It's not the relative values of say second and no outs versus bases empty and 1 out are a secret. If the value added by stealing X amount of times is outweighed by being caught Y amount of times, you're not adding value to your team. You can quibble around the margins of that equation if you like, but it's not like there's some super secret sauce that you can't see that's making it more valuable.

If you add in calculus that makes you think a guy might be a little more prone to injury, that just makes even less profitable to steal.

roger_that wrote: ↑Sun Dec 11, 2022 9:40 pm

Edgy MD wrote: ↑Sun Dec 11, 2022 6:21 pm I would disagree with that.

So someone who steals at a 75% rate, 30/40, is breaking even with 30 attempts (where he steals 20 bases) and netting you 10 bases on his other 10 attempts. How many runs is 10 bases? You know it's not 10 runs, not even close to that. Is it 5 runs? Doubt that very much. 3? 2? Something like that.

in 2022:

Stealing second with no outs, nets you 0.208 runs.
Getting throwing at second with no outs nets you -0.611

So if you steal successfully 3 times but get thrown out the fourth, you'll have earned .624 and cost .611. So you're up 0.013. So if you steal 231 bases at a 75% clip, you'll have earned your team one extra run.

Edgy MD wrote: ↑Sun Dec 11, 2022 6:21 pm I would disagree with that.

I think we've all read some variation on that, and allowed it to become an established truth, while we neglect to cite where we got it, and mostly don't remember where we got it, or what the break-even point actually is alleged to be.

Bill James has been researching the value of stolen bases at least since you were in elementary school, if not earlier. Tom Tango and then like a zillion others. The break even point is in the neighborhood of a 75% success rate and the math you'd need to calculate this yourself is fairly simple and straightforward. You just need to assemble the data so that you can then calculate break even. Or just trust James and Tango and like every other reputable sabrmetrician. They're not defending their break even figures the way others claim there's an invisible magician in the sky who runs the world. This is just math and logic.. There's no mystery here, not by any stretch of the imagination.

Break even might change radically in 2023 as the bases will be larger and pickoff attempts will be severely limited. Its gonna be easier to steal a base next season.

the game is better and more fun when stolen bases are in play.

unfortunately, the math says that unless you're really good at doing it successfully at the major league level, you're team is better off if you just don't.

at least, according to the maths that we are currently able to measure.

there may be secondary effects of having a base-stealer on first that are harder to measure, and it makes sense that these effects must surely exist - we just cannot measure how much value they have - of skewing the infield defense, or distracting the pitcher's focus, or degrading the catcher's ability to frame pitches, or affecting pitch selection, or being more primed to go first-to-third, or advancing on a bad throw, and so on and so forth - that i think teams and players should keep the SB in their arsenal, even before the upcoming rule changes, but it's probably a hard sell. i dont know how much those knock-on effects truly move the needle, but you're still going to need to be significantly more successful than not for it to be worth it.

If anybody wants to cite their studies, please do.

per a cursory google search:

https://batflipsandnerds.com/2018/11/03 ... olen-base/

just scanned through it a bit; has lots of math, draws some conclusions. seems pretty good.

i've been reading this kind of thing for at least the past 10-20 years, so there are plenty of other sources out there.

Thank you.

metsmarathon wrote: ↑Mon Dec 12, 2022 9:23 am

there may be secondary effects of having a base-stealer on first that are harder to measure, and it makes sense that these effects must surely exist - we just cannot measure how much value they have - of skewing the infield defense, or distracting the pitcher's focus, or degrading the catcher's ability to frame pitches, or affecting pitch selection, or being more primed to go first-to-third, or advancing on a bad throw, and so on and so forth - that i think teams and players should keep the SB in their arsenal....

Wouldn't these other effects already be baked into the formula, to a large degree? If these "secondary" effects impact run scoring, they would change the break-even number for base-stealing, which is a function of run scoring.

batmagadanleadoff wrote: ↑Mon Dec 12, 2022 7:57 am

Edgy MD wrote: ↑Sun Dec 11, 2022 6:21 pm I would disagree with that.

I think we've all read some variation on that, and allowed it to become an established truth, while we neglect to cite where we got it, and mostly don't remember where we got it, or what the break-even point actually is alleged to be.

Bill James has been researching the value of stolen bases at least since you were in elementary school, if not earlier. Tom Tango and then like a zillion others. The break even point is in the neighborhood of a 75% success rate and the math you'd need to calculate this yourself is fairly simple and straightforward. You just need to assemble the data so that you can then calculate break even. Or just trust James and Tango and like every other reputable sabrmetrician. They're not defending their break even figures the way others claim there's an invisible magician in the sky who runs the world. This is just math and logic.. There's no mystery here, not by any stretch of the imagination.

Break even might change radically in 2023 as the bases will be larger and pickoff attempts will be severely limited. Its gonna be easier to steal a base next season.

Why would larger bases change the break even rate? they would likely change the success rate for many players, thus changing the number who should attempt to steal more often as they are more likely to hit that break even rate, but why would the break even rate change?

the break even point will change because MLB will fuck with the baseball again. The bigger bases and the inane pick-off rule will have some effect, but it's hard to tell what exactly yet. It might be that the huge lead you'd be able to get if you can draw 2 throws cuts into a lot of the value of a SB, because if you can be 40% of the way already..

If you really want to get in the weeds with it, you could adjust based on park and offense. If you're a light-hitting offense in a low-run park, your break even point is lower. Same if you're playing against the Yankees and Aaron Judge, because MLB has made sure the balls are more likely to fly out of the park in those games. More chance of a HR means less difference between being on first or second.


As for secondary effects? we can measure those too. We have all sorts of splits for pitchers, runners on base, speed of runner on base. We can isolate out pitcher stats when Rickey was on base versus not. I don't know those conclusions off the top of my head, but they're not significant. You'd also have to look for potential batter distraction too. Perhaps rhythms get broken up for batter's too, when there's throw overs. This might be ESPECIALLY true in 2023, with the pitch clock. The pitcher still has quite a few ways to delay the game, and mess with timing. And with a faster pace of play and the pitcher still in control of it, it might mess with batter's more. We're absolutely going to get pick-off attempts on comically slow catchers, just so the pitcher can take an extra 15 seconds or whatever. Pete Alonso takes the lazy 40mph pick off throw, puts the tag on the runner in a comically slow manner. Looks at the ball. Adjusts his grip, Lobs it back softly.

batmagadanleadoff wrote: ↑Mon Dec 12, 2022 5:55 pm

metsmarathon wrote: ↑Mon Dec 12, 2022 9:23 am

there may be secondary effects of having a base-stealer on first that are harder to measure, and it makes sense that these effects must surely exist - we just cannot measure how much value they have - of skewing the infield defense, or distracting the pitcher's focus, or degrading the catcher's ability to frame pitches, or affecting pitch selection, or being more primed to go first-to-third, or advancing on a bad throw, and so on and so forth - that i think teams and players should keep the SB in their arsenal....

Wouldn't these other effects already be baked into the formula, to a large degree? If these "secondary" effects impact run scoring, they would change the break-even number for base-stealing, which is a function of run scoring.

the secondary effects can absolutely be measured, it's just that, to my knowledge, WE don't have ready access to that dataset.

they're not currently in the formula, because it looks at the entire population of base/out situations, not the quality of the runner or the expected outcome.

you would need to at minimum parse it out to the effect having a likely basestealer on first vs an unlinkely basestealer, and look at the change in the outcomes. there's a lot of small sample size things, and apples & orangutans comparisons that would come out, where with a narrower dataset you're probably going to need to normalize for the quality of hitter - presuming a better hitter could be more likely to follow a good basestealer and a less-good hitter may likely follow a worse basestealer. but given enough data, i'm sure it could be done.

give me a well-paying job, and maybe some python lessons, and i'll get right on it LOL.

but i do believe the secondary effects would change the break-even number. i don't think it's a wholesale change, but maybe a few percentage points of success rate, and maybe that's enough to encourage the entire volume of basestealing to rise, if the data were to bear it out.

also, it may be true that if mlb continues to deaden balls, the cost of an unsuccessful stolen base goes down. i think too that perhaps as k-rates increase, maybe advancing that extra base could become more valuable (?). math must be done! to the calculators, baseball nerds!

Changing topic slightly, Sherman said the Mets were asking the DBacks about Alek Thomas before they re-signed Nimmo. AZ asked for Baty in return (kinda comparable minor league numbers!) but the Mets decided to keep Baty and just buy back Nimmo because money isn't real for Cohen. Considering an alternate universe with Thomas the center fielder of the now and future and Escobar/Vientos at third for this year is kind of a fun exercise. Wonder if they woulda been content with Vientobar? Or try for Justin Turner for a year?

Thomas is an interesting thought. It's Sherman so I won't put too much thought into how real that "plan" was but looks like Thomas struggled with MLB, though still a good center fielder. I could see how you could think he would regain some of the BB% he had in the minors, and be a pretty useful MLBer. Though not as good as Nimmo, and the Mets offense is already a little hit or miss, and could probably use some more power. Still could use some more power. Maybe they have something in their three? prospects that on the cusp here, I dunno. Gotta try to find spots for them too, be interesting to see how that shakes out.

nymr83 wrote: ↑Mon Dec 12, 2022 8:47 pm

batmagadanleadoff wrote: ↑Mon Dec 12, 2022 7:57 am

Edgy MD wrote: ↑Sun Dec 11, 2022 6:21 pm I would disagree with that.

I think we've all read some variation on that, and allowed it to become an established truth, while we neglect to cite where we got it, and mostly don't remember where we got it, or what the break-even point actually is alleged to be.

Bill James has been researching the value of stolen bases at least since you were in elementary school, if not earlier. Tom Tango and then like a zillion others. The break even point is in the neighborhood of a 75% success rate and the math you'd need to calculate this yourself is fairly simple and straightforward. You just need to assemble the data so that you can then calculate break even. Or just trust James and Tango and like every other reputable sabrmetrician. They're not defending their break even figures the way others claim there's an invisible magician in the sky who runs the world. This is just math and logic.. There's no mystery here, not by any stretch of the imagination.

Break even might change radically in 2023 as the bases will be larger and pickoff attempts will be severely limited. Its gonna be easier to steal a base next season.

Why would larger bases change the break even rate? they would likely change the success rate for many players, thus changing the number who should attempt to steal more often as they are more likely to hit that break even rate, but why would the break even rate change?

The rules changes will likely change both figures -- the break-even point and the success rate. The two figures are independent of each other. All things being equal, base stealing will be easier in 2023. But this doesn't necessarily mean that the success rate will go up. because "all things being equal" is the operative phrase and all things might not be equal. The rules changes might also encourage lesser skilled base stealers to make more SB attempts, which could offset the improved rates that one would expect to be generated by the better base stealers. We'll see.

batmagadanleadoff wrote: ↑Tue Dec 13, 2022 8:43 pm

nymr83 wrote: ↑Mon Dec 12, 2022 8:47 pm

batmagadanleadoff wrote: ↑Mon Dec 12, 2022 7:57 am

Bill James has been researching the value of stolen bases at least since you were in elementary school, if not earlier. Tom Tango and then like a zillion others. The break even point is in the neighborhood of a 75% success rate and the math you'd need to calculate this yourself is fairly simple and straightforward. You just need to assemble the data so that you can then calculate break even. Or just trust James and Tango and like every other reputable sabrmetrician. They're not defending their break even figures the way others claim there's an invisible magician in the sky who runs the world. This is just math and logic.. There's no mystery here, not by any stretch of the imagination.

Break even might change radically in 2023 as the bases will be larger and pickoff attempts will be severely limited. Its gonna be easier to steal a base next season.

Why would larger bases change the break even rate? they would likely change the success rate for many players, thus changing the number who should attempt to steal more often as they are more likely to hit that break even rate, but why would the break even rate change?

The rules changes will likely change both figures -- the break-even point and the success rate. The two figures are independent of each other. All things being equal, base stealing will be easier in 2023. But this doesn't necessarily mean that the success rate will go up. because "all things being equal" is the operative phrase and all things might not be equal. The rules changes might also encourage lesser skilled base stealers to make more SB attempts, which could offset the improved rates that one would expect to be generated by the better base stealers. We'll see.

Maybe we are talking about different things here?

When I say "success rate" i mean the percentage of stolen base attempts that are successful. So if a player attempt 5 steals, steals 4 bases, and is aught once, his success rate is 80%. I agree that this number should go up under the new rules.

When I say "break even rate" I mean the rate at which a player must be successful for his stolen base attempts to be a net positive impact rather than a net negative impact on his team';s run scoring probability. Why would this change in the rules change that number?

nymr83 wrote: ↑Wed Dec 14, 2022 7:45 am

Maybe we are talking about different things here?

When I say "success rate" i mean the percentage of stolen base attempts that are successful. So if a player attempt 5 steals, steals 4 bases, and is aught once, his success rate is 80%. I agree that this number should go up under the new rules.

When I say "break even rate" I mean the rate at which a player must be successful for his stolen base attempts to be a net positive impact rather than a net negative impact on his team';s run scoring probability. Why would this change in the rules change that number?

We're talking about the exact same things. Break-even changes every year, albeit very slightly usually. Otherwise, by your logic, break-even would be a constant, a fixed number. Generally, break-even rises when run scoring is more plentiful. The easier it is to score runs, the more costly a failed stolen base attempt is.

BTW there are different break even rates for stolen bases in the same season. So for example, in the bottom of the ninth inning in a tied game, when the home team needs to score just one run to seal a victory, the break-even rate is lower. Under other circumstances, the break even rate would be higher because a failed stolen base attempt could potentially cost that team more than one run. But in that ninth inning situation, those extra runs are moot.

And that's a big part of why I registered my tendency to disagree. While a homerun or double may happen whenever, a stolen base is a strategy targeted to a specific situation, and so taking it out of the game situation and declaring it to be worth X runs which equals Y wins isn't particularly accurate.

when you put it in the game situation, you're still ascribing it to "worth X runs which equals Y wins", just with a more narrow focus.