Modeling Election to the Major League Baseball Hall of Fame through the use of Genetic Algorithms (PDF) by David Cohen
Here's the abstract:
This paper will use an alternate methodology for modeling called Genetic Algorithms. Using that method, several logical, rather than mathematical, rules for election to the baseball Hall of Fame will be found and examined. Predictions about future election, as well as past elections will be made. Ultimately, one rule will be picked as best, and examined in more depth than the others.
And the ultimate conclusion - For position players, you must meet at least 14 of the following criteria.
| Games | >2794 |
| AB | >3967 |
| Runs | >1145 |
| Hits | >1297 |
| TotalBases | >4432 |
| Doubles | >252 |
| Triples | >14 |
| HR | >355 |
| RBI | >325 |
| StolenBases | >341 |
| BB | >658 |
| HBP | >9 |
| Strikeouts | <2087 |
| Errors | <399 |
| NumAllStars | >9 |
| Psbb | >1 |
| Psstrikeouts | <42 |
| Psslugging | >0.269 |
| PSOBP | >0.010 |
Read the full paper - it's fascinating. And to my knowledge, matching fourteen of the nineteen criteria above does indeed lead to Hall of Fame election. There are no eligible position players that both (a) meet the criteria and (b) are not enshrined in Cooperstown.
So, who is currently playing or recently retired that will be elected to the HOF according to these rules? Who's just missing the cut? These are the kinds of questions I wanted to answer before I posted the article, but I simply haven't found the time. Now, I'm hoping that you can help. Run recent players through these rules and post your results somewhere - or email me at andrew AT andrewkoch DOT com and I'll post 'em for you.
And by the way, Dave Cohen is doing a great job in our Marketing dept...
