PITCHING STAT FORMULA MISTAKE

[atg8796]

I do stats for teams at the high school level. I was recently going through how iScore calculates every single stat because I wanted to create my own stat sheet on Microsoft Excel. I would figure out how iScore calculated a particular stat and then research the same formula on many different sites, including MLB's website to make sure iScore's formula was correct. There was one stat so far where I may have found an issue. For pitchers, there is a run average and an earned run average, commonly known as RA and ERA. ERA is calculated as total number of earned runs divided by innings pitching multiplied by 9, the 9 representing the fact that baseball is a 9 inning game. In my case, my stats have that ERA and an ERA where the 9 is a 7 instead because games are 7 innings at the high school level. In fact, all my stats are calculated for 7 inning games. Now on to RA... Run average is almost identical to ERA. The difference is that the total number of earned and unearned runs are used in the formula instead of just earned runs. With that being said, below are example numbers of how they should be done and iScore calculates them. You should see that iScores way is incorrect.

Total Innings Pitched = 50
Total Unearned Runs = 15
Total Earned Runs = 10
Total Allowed Runs = 25

CORRECT WAY
9 inning bases ERA = 10 divided by 50 equals 0.2, 0.2 multiplied by 9 equals a 1.8 ERA
9 inning bases RA = 25 divided by 50 equals 0.5, 0.5 multiplied by 9 equals a 4.5 RA

I-SCORE WAY
9 inning bases ERA = 10 divided by 50 equals 0.2, 0.2 multiplied by 9 equals a 1.8 ERA
9 inning bases RA = 25 divided by 50 equals 0.5, giving you an RA of 0.5 (THIS PART IS WRONG)

What iScore does for the RA is cuts out the part where you have to multiply by 9, or 7 in my case. Unless a pitcher never gives up an unearned run, it's impossible for a pitchers RA to be lower than his ERA. You hear all the time about a pitcher having a great ERA but a bad record because of too many walks or defensive errors. Example: a pitcher could give up 5 runs in 9 innings, but only 1 of them were earned. He might get the loss or no decision but had a great ERA in that particular game.

[FTMSupport]

In iScore, RA (Run Average) is simply Runs Allowed / Innings Pitched. We are showing the average runs per inning pitched, not per game. If you want per game, yes, you can multiply by the number of innings.

Having the RA value be runs per inning allows you to compare the Run Average of a High School game (7 innings) to the Run Average of a College Game (9 innings) or even the young Little League games (6 innings) and you still get an Apples to Apples comparison of the average runs per inning.

If you want to see the formulas iScore uses for stats, you can look here:

http://data.iscorecentral.com/legend.ph ... formulas=1

[atg8796]

Makes sense. Thank you for the reply. I was going off of what I commonly saw RA defined as on many other sites.

I had a few other questions/thoughts.

In the batting stats, is it possible for you guys to add in "total stealing attempts" and "stolen base percentage"? In my opinion no stat is complete without knowing all possible total sums or percentages. Like if you knew how many hits a guy had but didn't know how many total at-bats he had you'd have no way of figuring out his batting average.

In the fielding stats, can "total chances" be added? It shows the fielding percentage, how many errors, the types of errors, and how many put outs but not the total number of chances. Unless there's something I don't know about, the only way to figure out a single players total number of chances would be to go through the play-by-play of every game he was in to see how many chances he had. Once you knew how many chances he had you could compare that number with his total errors to figure to his fielding percentage. Example: in a game a player made 1 error and finished that game with a .900 fielding percentage. Unless you look at the play-by-play of that game you would have no way knowing if that player succeeded on 9 out of 10 chances or 18 out of 20 chances because either way their fielding percentage would be the same.

[FTMSupport]

Total Stealing attempts = bat_cs + bat_sb

Stolen Base Percentage = bat_sb / (bat_cs + bat_sb)

Chances = fld_putouts+fld_assists+fld_errors (this is the divisor of the fld_fielding_percentage stat in the previous link above)


The raw data is all there, so you can include these in your Excel spreadsheet if you would like.