For Stats Guys

[The Turtle]

Everyone is always looking for the next formula to rate hitters and some get quite complex, but lately I have been thiunking about all of the problems and I think I have decided that the way I best like to rate a player's season or career is to use this simple formula .. I think it best evaluates how bases were created in a year by one player with out getting too complex

TB+BB+SB+HPB+SH+SF-CS-GPD/AB+BB+HBP

so take for instance the debate of AL MVP this year A-Rod vs Ortiz

A-Rod had 369 TB + 91 BB + 21 SB +16 HBP +3 SF -6 CS - 8 GDP = 486 /
605 AB +91 BB +16 HBP =712

Total Base Average = .682

Does this percentage not say he will generate .682 bases per AB ???

Now take Ortiz

363 TB + 102 BB + 1SB +1 HBP + 9 SF -13 GDP = 463 /

601 AB + 102 BB +1 HBP = 704

Total Base Average = .658


Am I off here saying this is a good way to rate players?

Has someone already devised this formua? (I'm sure they have)

And is this not a simple way to rate players and more effective than OPS???

[1crazycanuk]

I don't understand why you got HBP and BB's in both sides of the equation. I'm not a math whiz but I'm thinking that wouldn't work?

[The Turtle]

Well, to put it simply they are on both side in terms of OBP which is measured simply by H+BB+HBP/AB+BB+HBP

[maligned]

This is the formula for Runs Created that almost perfectly predicts how many runs any team will score. If you take an entire season of stats recorded by a league, this formula will perfectly predict how many runs were scored.

[(H + BB + HBP - CS - GIDP) times (Total bases + .26[BB - IBB + HBP] + .52[SH + SF + SB])] divided by (AB + BB + HBP + SH+ SF)

If you want to apply it to an individual player as a rating system, you can simply divide the value by plate appearances to give you RC/PA or better yet, like strat does, divide it by the outs he recorded, then multiply by 27 outs (the number in a full game) to tell you how many runs a team of this individual would score per game. This is the stat that appears in the Strat individual stats list: RC/27.

The philosophy behind this stat is simple: OBP * TB = Runs Created. Math wizzes with more time on their hands than me have created the above formula to include the details you tried to include in yours: effects of SB, BB, HBP, CS, etc. The problem with your formula is that SBs, BBs, etc. have less "base-earning" value when predicting runs created than actual singles, double, etc. That's why this formula multiplies BB and HBP by .26 and SB, SF, and SH by .52.

Let me know if any of that didn't make sense.

[The Turtle]

I'm very familiar with that formula, but I am looking for something simple, easy to explain and does not depend on linear weights for how much they are worth and can be used by people much like OPS, which has many flaws

I just do not see the general public picking up that stat due to those weights that they do not know how thye were derived - although I do like it and genrally feel the MVP should go to the player with the most RC on a division winning team just as Arod had about 151 RC while Ortiz had 140

Basically what I am asking is if anyone has seen an easier formula that determines how productive a hitter can be and does not depend on someone being right about a walk or a sb's worth relative to a single

[maligned]

Gotcha. I agree that yours is more accessible.

Another idea would be to just do a simplified Runs Created per Plate Appearance: (OBP*TB)/PA

Many stat lines contain these three stats, and you don't have to screw around with the small bonuses or penalties for GDP, CS, SB, BB, etc.

Another thought I've had is this: why didn't whoever created OPS at least multiply the two values instead of divide? This at least gives you RC/AB, instead of the flawed value given by adding OBP and SLG.

[The Turtle]

Thats exactly the thoughts I am looking for here, maligned, something easy and better than the doubling issue that OPS gives us, why exactly do we use OPS when we know it has very visible flaws but is not any easier than OBP*TB/PA or even TB+BB/PA

[The Biomechanical Man]

I see what you are getting at, Turtle. Your formula is bases created per plate appearances:

[(Bases created) - ("extra" bases lost)] / (Plate appearances)
[(TB+BB+SB+HPB+SH+SF) - (CS+GPD)] / (AB+BB+HBP)

The "extra bases" are above and beyond the standard out at-bat.

[The Turtle]

Exactly, Glen, my average is just saying this is how many bases a player is going to get every time he steps up to the plate, although when i look at it again I should perhaps take out GDP and SF/SH as this is a result of others being on base and not the individual player at bat, but these are minute changes...SB and CS can be a larger factor especially if you are looking at Carl Crawford, Juan Pierre, or especially Vince Coleman/Rickey


btw the highest rated season I can find using my base average is Bond's 2001 season at .909

[PAULMINICUCCI]

I like the formula you proposed. It is direct and simple. Speaking to the last point aboutSF being the result of someone on base and not due soley to the efforts of the batter, neither are IBBs because the opposing manager for tactical purposes is making the play, not the batter. I really think the RC that uses the linear products is misleading for that reason, it actually has a slight disadvantage to Bonds in that if Bonds was able to bat the 70 or so IBBs he would generate more bases than by being walked. Now that I think about that though he probably would have to have a SLUG of 1.000 to be better than an IBB for himself, although an IBB will almost never (I saw Bonds get walked on an IBB with the bases loaded once) generate 2or 3 or 4 bases with one swing.

The only other thing I would say is this, I think this game and others understate the value of speed as a part of creating runs, in that a runner trying to steal causes the defense to have to make a play they wouldn;t otherwise make, and the value of having to hold on a runner at first that gives the LH hitter an advantage (which I realize strat takes care of with the * for stealers), nonehteless, creates a base (the single the batter now hits to right or the force play instead of the doubleplay) taking extra bases adds to stress on the defense and does result in extra bases from miscues. That kind of tension, the Rickey factor if you will ( I know Rickey would call it the Rickey factor if Rickey had the slightest idea of what we are talking about).

All in all, I think your formula works well and does do justice to the Ortiz A-Rod debate. Thanks for passing it on.