Stolen Bases – How They Workin’?

[RonCo]

After the discussion in the 100+ SB Poll, I did a quick pull of our league’s data to get a read on how various players might be expected to do. Given sample sizes, there’s going to be wide variance, but if you take a look at certain guys, the following charts might give you a few insights on things. So, in the spirit of shared learning, and with a gazillion caveats having to do with YMMV and all that good stuff, here’s a breakdown.

Method:

1) In OOTP, go to Stats/Sortable Stats
2) Create a view that has SPEED/STEAL/SB/CS in it.
3) Write the report to Disk (opens it in a browser)
4) Copy the data to Excel
5) Do a quick pivot table with SPEED and STEAL on the row and columns
6) Have the pivot table calculate the values for SB, copy those off in another worksheet
7) Have the pivot table calculate the values for CS, copy those off in that same worksheet
8) Plugin a few quick equations to do the adding and dividing.

Total Time: Maybe five minutes (takes longer to describe it than do it. )

What you’re shooting for is a pair of tables, one that shows how often players of each Speed/Steal rating-pair have attempted to steal in BBA games so far (for example, of the 7,220 attempts so far this year, 2,983 [41%] have been attempted by players currently rated 10/10. This is interesting to think about). The second table is identical to the first, except that it shows success rates.

Here are Major Things That Influence Outcomes that are not in the chart:

• Pitcher Hold Ratings
• Catcher Arm Ratings
• Stealing of 2B vs. 3B
• Failed Hit and Run
• Players who have changed ratings over the season will be in the “wrong” bins

It’s also possible that Base Running Ratings and Intelligence impacts results at some level.

Bottom line: this is just a simple SPEED/STEAL Chart. It’s easy to use, but possibly wrong!

Some things that jump out at me:

1) The SPEED Rating does not seem to influence success much at all. This is consistent with other versions of OOTP to my knowledge. The STEAL rating does influence it in expected fashions.
2) The SPEED Rating appears to have a much stronger influence on how often players actually run. This makes sense. Bottom line: 5,233 of the 7,220 attempts (72%) have been with runners rated 10 for SPEED. 3,560 attempts (49%) have been made by players with STEAL ratings of 10.
3) By looking in OOTP’s historical analysis page, it’s clear the game is attempting to keep the success rate pegged at 66%. This is maybe a little low, but whatever…the league’s current success rate is 66.7%. Think about what this means when you look at what players run most often (72% have SPEED=10, 49% have STEAL = 10). If the league is going to steal at 66%, and mostly fast runners run, the CS have to go somewhere. I fully admit I don't know how the game really does this, though.
4) Take some scans up and down the table. The data has sample size noise that you can assess yourself, but it tells me stories. I’m not sure what to do with these stories, but they are stories nonetheless.

[RonCo]

Like I said, I don't really know what do do with this data all the time, but I like to think about it. For example take Yellow Spring's Angel De Costillo.

Speed = 10
Steal = 4

Per this chart there have been 53 stolen base attempts by 10/4 guys, and they have been successful 40% of the time. I can tell you this is interesting because Castillo has been the instigator of a whopping 38 of those 53 attempts. I've used the game strategy tabs to tone him down, but he still attempts to steal fairly often.

Is this the manager? (His ratings say he doesn't like to steal a ton). Is it game situation (I do try to run at certain times)? Is it Hit and Runs? (Again, Richards theoretically uses that strategy fairly rarely).

I don't know.

All I can say for sure is that De Castillo has run 38 times, which is rare for guys with his ratings.

[Lane]

wow. that's excellent. really valuable information here that will likely impact my strategy sliders.

[Lane]

it also makes me even more interested in seeing the ratings over 10...I wonder how the rates change over 10, though I suspect sample size will be too small over one season.

[udlb58]

That is pretty much in line with my experience....

Speed = how often a guy will attempt to steal (and I think it has some affect on triples as well, maybe combined with Baserunning?)
Steal = how successful the guy is when attempting to steal
Baserunning = ????? I'm not really sure other than his ability to successfully execute a hit and run on the bases or to stretch extra bases out of balls put in play

So I usually crank the steal slider way down for guys with high speed and low steal, while moving it up a little more than normal for someone with high steal and low speed.

Overall, this is a notice to basically expect 60% or less from guys with 9 steal or lower (Don't run heavily with guys unless they are 10 steal!). Because of the way OOTP tries to hit set statistical parameters, with a high number of 10 steal players, we basically end up with only 10 steal players being able to succeed at an acceptable rate.

[Lane]

yeah, this screams "don't steal with guys unless they're 10 steal"

[allenr]

Pretty much. And it seems kind arbitrary which 10/10 guy gets the CS. Look at Alain Pichon last year vs this. He's been 10/10 both years, but he got absolutely destroyed with CS last year. And in 2029.

[RonCo]

Like I say above, be a little careful with making generalizations here because there are a lot of moving parts under the hood. One, for example, is that good catcher arm tends to keep slower runners from attempting to steal... Or, to put it another way, fast runners are the primary guys who run at disadvantage.

Also, there is a glut of fast guys in the league. Do your own data cut for the exact numbers...I suggest this has some overall effect, too. But I couldn't say how to use this idea in the sliders beyond just sitting them off... Which can make your team less effective in certain high leverage moments.

[udlb58]

Like I say above, be a little careful with making generalizations here because there are a lot of moving parts under the hood. One, for example, is that good catcher arm tends to keep slower runners from attempting to steal... Or, to put it another way, fast runners are the primary guys who run at disadvantage.

Also, there is a glut of fast guys in the league. Do your own data cut for the exact numbers...I suggest this has some overall effect, too. But I couldn't say how to use this idea in the sliders beyond just sitting them off... Which can make your team less effective in certain high leverage moments.

Or, maybe that we just need to pay a little more attention to SB%. Despite being 10/10/10, Chavez has never had a high success rate, I should probably send him less. Though he's 10/10/10; his success rate the last 3 years hasn't really been better than Medrano and his 6/8/10 ratings.

[allenr]

Which is really silly if you ask me what's the point of ratings at that point? What does haves scout page say related to baserunning?