NERP and Catcher Defense

you're a {content removed for uncertainty} barrmorris--thanks

As for the answer. It is 42.

Thanks Frankie, but how do you know I'm a man?

I've added 10 simulations to each configuration (for a total of 30) - updated results are below:

Arm	SB Strategy	Avg. SB	Avg. CS	Avg. SB%	Avg. Runs Allow.
-5	Normal	33.9	14.3	70.4%	882
+5	Normal	120.8	33.6	78.3%	871
-5	Ex. Conserv.	21.1	7.5	73.7%	870
+5	Ex. Conserv.	86.2	17.9	82.8%	865

Now, none of the runs allowed differences are statistically significant. I think the most we can say regarding the impact of catcher's arm on runs is that whatever direction the impact is, it is small in magnitude.

ha--well it is a gender neutral expression in a way that "you guys" is not literal either.

Never said I know anything--after all, I don't really know you are good either.

I can definitely agree with that - I am a {content removed for uncertainty}

the double entendre(?) on that is pretty funny

the double entendre(?) on that is pretty funny

I saw a double entendre used in one of those movies once.

But seriously, to everyone that's added info and insights on this thread, thank you.

Results of more simulations. Still 2021, Realmuto with adjustments to range. 40 sims each. I decided that using confidence intervals (CI) might express uncertainty better.

Range	Blocking ?	Mean	80% CI	95% CI
1	No	874	+/- 52	+/- 93
5	No	886	+/- 57	+/- 93
1	Yes	867	+/- 45	+/- 69
5	Yes	888	+/- 50	+/- 76
Range	Blocking ?	Diff.	80% CI	95% CI
1 vs 5	No	-11	+/- 8	+/- 17
1 vs 5	Yes	-20	+/- 7	+/- 15
1	Yes vs no	-7	+/- 7	+/- 15
5	Yes vs no	2	+/- 8	+/- 17

We know that given how blocking the plate works, it should be symmetric (magnitude of runs reduced by a 1 should equal magnitude of runs given up by a 5), so from this, my point estimate of the impact of blocking the plate would range from -4.5 to +4.5 for catcher's range from 1 to 5. I think Marc estimated it at -5 to +5 (going on memory, I can't find the post), so that looks pretty good. The estimate of the x-chart difference of 11 runs between a 1 and a 5 is close to my estimate of 13 based on analysis of the x-chart.

So, on this team of mine, my lefty platooner Smoky Burgess (+1) has given up 20 steals, but caught 18 runners for outs.
Whereas my righty platooner Steve Lake (-2) has only given up 3 steals, but only caught 2 runners for outs. Burgess is an outmaking machine because runners take a chance on him.

https://365.strat-o-matic.com/team/fielding/1727728

Which is better in terms of defensive runs saved etc? (Or whatever yardstick to measure it by.) Burgess accounted for 18 outs, that's a lot. Whereas Lake only got two runners out. But 17 more runners advanced at least one base on Burgess.... I don't have a data-driven answer (but my gut leans toward Burgess being worth it). Just wondering if others have insights.

Oh I didn't even notice that result lab pointed out. Seems bad from a game design standpoint...HAL's SB logic is clearly borked if weaker arms are preventing more runs than stronger arms.

You can't conclude that from this data. The differences are not statistically significant which means that there is some chance that they are samples from a distribution with the same mean. The difficulty in measuring the impact of a small factor (like catcher's arm) on runs is that the standard deviation of runs is around 40. I'm skeptical of the results for runs allowed by arm strength with Extra Conservative SB strategy. Other analysis I've done of run values suggests that success rates this high should produce positive value for SBs. I may run more simulations, but this is a tedious process.

Thank you so much barrmorris and all! Childsmwc especially!

I'd be happy to run more simulations on PC (I do that often), though haven't quite pinned down what variables are different and how critically they are---and how to best "align" the PC game to SOM365 play. You might need to guide me through your methodology but I love running/saving data from sims I understand (a bit) of what you're saying about the terrible throwers being "better" not being statistically significant, but it still stands out as significant. I'd love to see if that keeps holding true or what might be accounting for it.

About to dive back in to this research and hopefully refine my previuos refined numbers, with just a few more insights on the more speculative parts I outlined above-ish-somewhere.

So, on this team of mine, my lefty platooner Smoky Burgess (+1) has given up 20 steals, but caught 18 runners for outs.
Whereas my righty platooner Steve Lake (-2) has only given up 3 steals, but only caught 2 runners for outs. Burgess is an outmaking machine because runners take a chance on him.

https://365.strat-o-matic.com/team/fielding/1727728

Which is better in terms of defensive runs saved etc? (Or whatever yardstick to measure it by.) Burgess accounted for 18 outs, that's a lot. Whereas Lake only got two runners out. But 17 more runners advanced at least one base on Burgess.... I don't have a data-driven answer (but my gut leans toward Burgess being worth it). Just wondering if others have insights.

Insights, no.
But an anecdote of the arms and assists from real life.
I played LC with the same guy playing LF for maybe 10 years. A lot of the opponents were the same year after year.
I covered a lot of ground but had a noodle arm. From medium depth where I'd regularly position myself, the SS needed to come maybe 1/3 of the way out to get a throw from me. If I had to go back, he needed to take 3 steps for every 1 I took.
You get it.
Guy in LF could throw home on a fly if he could step into it, hit 2nd and 3rd flat footed.
He's a 3 (-4)
I'm a 1 (+3) or worse.
I had more assists than him every year. I can't quantify anything beyond that, but when a hitter rounded first and I had the ball, he night go. If the LF had the ball, he probably wouldn't. First to third, they'd run on me a lot. I had more chances at assist by a large stretch, and got some of them.

I was Smokey Burgess!