Single Best Offensive Statistic

[Second Base Coach]

My favorite review of Stat One (from a Blog) includes this:

Messmer states at the end of Sisler’s entry, “He had negligible power, however, and that prevents him from placing in the top five for this position. He does make the top 10, though” (pg.64). Fine. Let us check where he falls in the Top 10 first basemen. Here is the list in descending order. Jim Bottomley, Bill Terry, Eddie Murray, Johnny Mize, Mark McGwire, Jeff Bagwell, Albert Pujols, Hank Greenberg, Jimmy Foxx, and Lou Gehrig. George Sisler does not appear on the list. You can look up later that George Sisler is the 74 best player of all time and Jim Bottomly the 71st. So, despite the comment that Sisler makes the Top 10, he does not.

[Second Base Coach]

Did you know that StatOne is multi-media?

http://video.aol.com/video-detail/st...60?icid=acvsv3

[RuthMayBond]

On a related note, you can take some pretty awful stats and still come up with a decent top list of the best second basemen of all time.

On another related note, how can you take some ultimately flawed stats and come up with one that "is not"?

[Second Base Coach]

After looking at P/E... I could not help but notice his mismanaged attempt to use caught stealing. I posted this on youtube for him to read:

Just looking at your stat, you have made a mistake.

You are using caught stealing, good for you, but you should not be using it to take away from the stolen base number (in a sense you are using a stat called Net Steals).

That Eff part of the numerator is doing this then... adding bases, then subtracting.... wait for it... not a base but subtracting outs from the total number of bases.

This is a huge apples and oranges mistake. The outs he makes on the bases is much worse than the stolen base he creates for his team.

And you might want to consider something else besides plate appearances. How about outs, or something.

If you don't believe me, then consider what you would add in to the Eff part of your stat if a player doubled, stole third, then got caught stealing home.

To lead off the game... He cost his team an out, yet your stat would give him credit for the double, then add in a steal, then subtract it away. Your stat would give the same credit to a player who did that compared to a slug who just stole second and sat there for the whole inning.

[RuthMayBond]

My favorite review of Stat One (from a Blog) includes this:

Messmer states at the end of Sisler’s entry, “He had negligible power, however, and that prevents him from placing in the top five for this position. He does make the top 10, though” (pg.64). Fine. Let us check where he falls in the Top 10 first basemen. Here is the list in descending order. Jim Bottomley, Bill Terry, Eddie Murray, Johnny Mize, Mark McGwire, Jeff Bagwell, Albert Pujols, Hank Greenberg, Jimmy Foxx, and Lou Gehrig. George Sisler does not appear on the list. You can look up later that George Sisler is the 74 best player of all time and Jim Bottomly the 71st. So, despite the comment that Sisler makes the Top 10, he does not.

Pujols (who knows how long ago his book was written?) and Bottomley should thank him, Frank Thomas (I suppose he's pleading DH), Anson & McCovey not so much. I guess Anson & McCovey were no Bottomley

[Bowler Bob in Brisbane]

What do you think is the single best offensive statistic to judge hitters by?

I've become very partial to the TROBBI (Total Runners On Base Batted In) Percentage, which is total RBI's minus the batters total HR's divided by the total runners on base during the batters total plate appearances.

A good example is Eric Hinske's TROBBI PCT., which currently as of Thursday night June 26th stands at an even 20$%.

In the period starting March 25th through June 26th, roughly 79 games, Hinske had a total of 145 runners on base when he came to the plate, and had a total of 41 RBI's and a total of 12 home runs in that time period.

By subtracting his 12 total home runs from his 41 total RBI's that leaves 29 tota base runners delivered to the plate.

To get the TROBBI percentage divide the 29 total baserunners driven in by the 145 total baserunners, and you get 0.2, or 20%, which essentially means that Eric Hinske drove in 1 base runner for every 5 runners on base during his plate appearances.

and there you have it, the TROBBI percentage.

Cheerss!

BB in B

[Second Base Coach]

That is a pretty good number. The clutch hitting crowd would love that kind of stuff. Those against such numbers often put out the fact that when you look at a large number of players and you compare their overall batting average to their batting average with runners in scoring position, there is not much difference. And the good players seem to hit just a little better as a group. That stands to reason, as they are the good players, right?

I am not one to buy into clutch hitting. I blame that to the pitcher being on the ropes.

But that above stat is a good indication if a player is having a good year to date. (or not) I am not sure if we can rely on that to predict future results. For every stathead who claims a batter has been clutch so far in April May and June, I can find another stat head who will tell you that he will most likely fall back to his historic mean and you should sell high if you have him on your fantasy roster.

However, if he he stays hot (lucky?) all year long, then those RBI warriors tend to do well in MVP voting.

[RuthMayBond]

I've become very partial to the TROBBI (Total Runners On Base Batted In) Percentage, which is total RBI's minus the batters total HR's divided by the total runners on base during the batters total plate appearances.

A good example is Eric Hinske's TROBBI PCT., which currently as of Thursday night June 26th stands at an even 20$%.

In the period starting March 25th through June 26th, roughly 79 games, Hinske had a total of 145 runners on base when he came to the plate, and had a total of 41 RBI's and a total of 12 home runs in that time period.

By subtracting his 12 total home runs from his 41 total RBI's that leaves 29 tota base runners delivered to the plate.

To get the TROBBI percentage divide the 29 total baserunners driven in by the 145 total baserunners, and you get 0.2, or 20%, which essentially means that Eric Hinske drove in 1 base runner for every 5 runners on base during his plate appearances.

and there you have it, the TROBBI percentage.

Cheerss!

BB in B

It says nothing about which base the runners are on (not to mention how many outs). Bottom of the order guys and leadoff are much more likely to have fewer guys on and probably more likely on first, than the 3-4-5 hitters

[PVNICK]

It might be more useful if it was broken down into men in scoring position or something. It provides another facet of information on a player, whether it is reflective of an ability above and beyond that for the most part the leaders were also amogn the best hitters.

[digglahhh]

It might be more useful if it was broken down into men in scoring position or something. It provides another facet of information on a player, whether it is reflective of an ability above and beyond that for the most part the leaders were also amogn the best hitters.

Certainly.

Man on 3rd, w/ 0 out vs. man on first w/ 2 out - these are not the same situation, by any means.

John Walsh's, Confessions of an RBI Fanatic provides some interesting and more nuanced work on this subject.

[PVNICK]

John Walsh's, Confessions of an RBI Fanatic provides some interesting and more nuanced work on this subject.

Thanks for the info. There's some rather ironic data in there. I wonder if it breaks down by type/style of hitter. Somehow I doubt it.

[skyking162]

Fangraphs has run expectancy runs. Hitters are given credit/blame for the change in expected runs scored because of their plate appearances. It's like linear weights with the base-out leverage left in. Going one step further would be WPA. It all depends how much of the game context you want to matter and therefore what question you're asking.

[SABR Matt]

I'd rather use Run Expectancy runs than WPA as my tool for analyzing a hitter's productivity. WPA penalizes hitters for being on teams that have a lot of blow-out games or that are bad teams. RE Runs is a better tool for individual analysis.

Of course...Linear Weights is better than RE Runs IMHO.

[tip184]

To me, the most important hitting stats in order are:

1) Plate appearances per home run.
2) On base percentage.
3) Plate appearances per double.

Doubles and home runs cause completely different results, and should not be put together in the same statistic (slugging percentage). Doubles drive in 1,2 or 3 runners and/or put the batter in scoring position. Home runs automatically score one to four runs. Depending on whether or not there are runners on base, they end up producing different results in terms of run production.

[SABR Matt]

That may be the goofiest thing I've ever seen anyone write on this forum. Sorry dude, but doubles and HRs are no more different than outs and hits (and you made no objection to batting average). Of course it's true that doubles are less valuable than HRs...but then any good linear weight based offensive metric will see that (0.78 runs for doubles, 1.4 runs for HR).

People get such bizarre ideas in their heads sometimes...

[tip184]

That may be the goofiest thing I've ever seen anyone write on this forum. Sorry dude, but doubles and HRs are no more different than outs and hits (and you made no objection to batting average). Of course it's true that doubles are less valuable than HRs...but then any good linear weight based offensive metric will see that (0.78 runs for doubles, 1.4 runs for HR).

People get such bizarre ideas in their heads sometimes...

Well I'm sorry but I'm just throwing some ideas out there. Am I not allowed to do that?

[SABR Matt]

Of course you are. I am, however, allowed to comment on those ideas and express my opinion that they are illogical.

[PVNICK]

Fangraphs has run expectancy runs. Hitters are given credit/blame for the change in expected runs scored because of their plate appearances. It's like linear weights with the base-out leverage left in. Going one step further would be WPA. It all depends how much of the game context you want to matter and therefore what question you're asking.

Wow and I was going to ask or start a thread to see how much if anything was out there along those lines.

[weskelton]

Well I'm sorry but I'm just throwing some ideas out there. Am I not allowed to do that?

Tip,

Don't get discouraged. If this thread is any evidence, you might have the makings of your own book.

[Second Base Coach]

Call it "Stat Two", as in two-bagger. I would buy it, just don't subtract outs from bases...

[King Diamond]

Call it "Stat Two", as in two-bagger. I would buy it, just don't subtract outs from bases...

How about "STAT SUCKS" because that's what I think of that book.