Statistics That Matter
There are a lot of details on “statistics that matter” that need to filled in. I hope these show notes are useful.
In the podcast, I mentioned that Pythagoreom formula to calculate winning percentage based on average runs per game. This is really called the “Pythagoreom Expectation” for winning percentage. Click here for some background on this from Baseball Prospectus.
The formula for winning percentage is as follows:
Winning % = (Runs Scored)^2 divided by ((Runs Scored)^2+(Runs Allowed)^2))
This formula is incredibly accurate in projecting winning percentage at all levels of baseball.
At the team level, there are 3 statistics that matter:
Win 3 innings
Have 15 Quality At Bats
Limit Walks + Errors < 6
These are by far the three most important elements of winning baseball and teaching players the right way to play baseball. This is the holy grail of statistics.
Winning 3 innings teaches players to compete relentlessly, inning by inning, at bat by at bat and pitch by pitch. It is really hard to win 3 innings in a game.
There are 4 ways to have quality at bats.
1 - Hit the ball hard, regardless of outcome. All line drives are hard hits. Hard hits also include hard hit groundballs and flyballs that force outfielders to retreat on the ball.
2 - Draw a walk
3 - Execute - sacrifice, advance the runner to 3rd, get a run home with less that 2 outs, hit & runs, squeeze, etc..
4 - Any at bat that consumes 6 or more pitches is a quality at bat - regardless of outcome.
A quality at bat is worth between .4 and .5 runs. Using the Pythagoreom Expectation, if you want to increase your winning percentage from 50% to 67%, you need to score 1 more run per game. This means your team needs to produce just 2-3 more quality at bats per game. Now, that is something that players can focus on and control - IF they understand what a quality at bat is.
Finally, walks + errors are the defensive mirror image to quality at bats. When you walk batters and allow runners to reach base on errors, then you are giving away outs. Any time you give away an out - with a walk or an error - on average you cost your team .4-.5 runs.
So, if you want to reduce runs allowed by one run per game. Then, you need to reduce walks + errors by 2-3.
Those are the statistics that matter for the overall team. This is how you can teach baseball and win games at the same time.
For individual players, the statistics that matter focus on things that are more within the players control.
Here is what I use:
On-base percentage - this the only conventional measure that I care about. You need to get on base in order to score runs. A good target in youth baseball is >450.
Quality At Bats - this was explained above. The target should be >60% quality at bats, with at least 40% coming from hard hit balls. These numbers are hard to achieve. Most players will be at 50% and 30% respectively.
Walk:Strikeout Ratio - this is the only measure on pitch selection that I have found that can be used effectively in youth baseball (this is because there is no record of pitch location). If a player has good pitch selection, he should be swinging at strikes and taking balls. A player without good pitch selection takes strikes and swings at balls. I want players to walk 1.5x more than they strikeout. Strikeouts are OK, as long as the hitter is aggressively swinging at good pitches and laying off bad pitches. You’ll notice that players that are struggling will be striking out more than they walk. This is a pitch selection problem.
Runs Created is the ultimate measure of hitting production. It is a complicated formula that takes into account the relative value of different at bat outcomes. Click here and click here if you are interested in the background and details of how this formula is applied in professional baseball. If you want to use outcome based performance measures, then throw batting average out the window and use Runs Created.
Here are some articles on Runs Created - Article 1 and Article 2.
Over the years, I have made some adjustments to the Runs Created formula for youth baseball. Here is my version of the formula, based on a 6-7 inning game.
RC =((Hits+Walks)*(Total Bases + ROE) + .26(Walks + ROE) + .52*SAC)/PA
RC/6 innings = RC / ((PA-BB-ROE-H)*18)
RC/7 innings = RC / ((PA-BB-ROE-H)*21)
ROE = reach on error (in mlb, the hitter is not given credit for ROE. In youth baseball, a lot of errors are created by hard hit balls and hustling runners. So, hitters should get credit for ROE)
Walks = includes walks and HBP
PA = plate appearances
Again, Quality at Bats are by far the most important stat to use for youth players. If you must use a outcome based measure than use On-Base % and Runs Created. They are much more meaningful than batting average.
For pitchers, the statistics that matter are:
% Strikes - because we want pitchers to keep the game moving and force the other team to swing the bat.
Pitches per inning (3 outs) = because we want our pitchers to be efficient and protect their arms
Opposing team On-Base % - Calculated as (walks + hits)/batters faced
Runs Allowed per batter faced - what percentage of batters that the pitcher faces end up crossing the plate. This is a lot more meaningful that ERA for youth pitchers.
At the more advanced level, the following measures are important:
% First Pitch Strikes
% of batters that are on base or out in 3 pitches
% Offspeed strikes
Here are the target measures for the statistics that matter:
For hitters:
Good at bat %, Target >60%
OB % > .500
Walk:Strikeout Ratio >1.0
For pitchers:
% Strikes >65%
Pitches per inning <15
Runs Allowed per Batter Faces >.200
Opposing Team On-Base % >.350
First pitch strikes >67%
On or out in 3 pitches > 50%
Offspeed Strikes > 50%