The Run Value of Outs

by Graham MacAree on Jul 9, 2008 3:18 PM PDT

Today we'll have a quick look at what tRA can tell us beyond pitching.

Since calculating tRA requires expected outs and expected runs, we can utilise these figures to look at defense - in this case to check whether a missed play on defense is really worth the 0.8 extra runs we expect.

Determining xR-R and O-xO for the entire league... tells us nothing, because the resulting errors are tiny (which is good for tRA, but not good for what we're trying to do). Instead, we'll take the absolute value of xR-R and O-xO for each pitcher in, say, the AL last year, and sum that instead. We can then divide the two values and see how close we get to 0.8.

xR-R: 783.9926

O-xO: 978.7843

(xR-R)/(O-xO) = 0.80096

Well that's kinda neat.

Tweet Comment 21 comments | 6 recs |

Do you like this story?

Comments

In case people don't know, the value of 0.8 runs per out/play came from Tango.

And while, I don’t have the actual link anymore, here was his explanation of the concept:

Suppose a team with Ozzie at SS gives up on average 12 non-HR hits, and 2.6 walks every game (which of course is 27 outs). Applying .50 runs per non-HR hit (I know it should be closer to .57, but I just want to keep it basic), and .30 runs per BB, and -.10 runs per out, and we get 4.08 runs scored per game. And per game, we see that Ozzie’s team faces 41.6 batters (again, let’s not worry about DPs, etc).
Now, let’s say Ozzie was traded for Spike, and let’s say for every 41.6 batters faced, there is one ball that Ozzie gets to that Spike doesn’t. So, for those 41.6 batters, Spike’s team records 13 non-HR hits (1 more than Oz), 2.6 walks, and 26 outs (1 less than Oz). However, there’s still one more out to go! Since Spike’s team gives up 13 non-HR hits / 26 outs, we can estimate that this team will give up 13.5 non-HR hits, 2.7 walks, and 27 outs per game ( a total of 43.2 batters, a remarkable 1.6 MORE batters than Oz). Anyway, applying our LW constants, and we see that Spike’s team gives up 4.86 runs per game.

This number is .78 runs MORE than Ozzie. This is the result of Ozzie getting to one more hit than Spike. .50 runs for the hit, and about .30 runs for the out gives you the .80 runs.

by Matthew on Jul 9, 2008 3:25 PM PDT reply actions

This was the sum of abs(pt_1) + abs(pt_2) ... + abs(pt_x) correct?

and do you think there’s any value to trying to apply the tRA application to a non-ML pitcher?

by seattlebruin on Jul 9, 2008 3:27 PM PDT reply actions

Yes to both

You’d just need to adjust run and out values for different leagues if you wanted to apply tRA to the minors, and that’s tricky because PBP data is hard to come by

by Graham MacAree on Jul 9, 2008 3:29 PM PDT up reply actions

I think it's possible

Just get whatever Sackman uses/used to create minorleaguesplits and use that database for this purpose.
The tough thing would be coding the bot to grab the PBP stuff from MiLB.com’s ‘recaps’ – but in this case, I think that’s already been done.
They MAY have run values over there, though I didn’t see it on a quick look a year or two ago when I was trying to create WPA charts for the Rainiers… may have been added, though.
Out values…yeah, good luck with that.

by marc w on Jul 9, 2008 3:33 PM PDT up reply actions

We know how to calculate everything from scratch if it comes to it

by Graham MacAree on Jul 9, 2008 3:34 PM PDT up reply actions

Oh, I know.

Just hoping there are shortcuts is all….

by marc w on Jul 9, 2008 3:44 PM PDT up reply actions

I think Matthew is able to automate it all given the PBP data anyway

by Graham MacAree on Jul 9, 2008 3:46 PM PDT up reply actions

I think that's one of the job requirements for data ferries

by Last Fan Of Jose Lopez on Jul 9, 2008 3:48 PM PDT up reply actions

Yeah, that's what I'm worried about

I was going through the 23K game and discovered the SDSU actually has a play by play of every single game, which would allow you to get a fairly good look at Strasburg’s LD/GB/FB/IFB ratios – obviously you’d be guessing on the hits, but you could probably re-create some of that from the recaps. The only thing is that you would have to re-create the run values for (probably the MWC), which would obviously be very very difficult.

by seattlebruin on Jul 9, 2008 3:34 PM PDT up reply actions