Major league baseball uses the Elias Bureau’s ratings to determine type A and type B free agents. (If you have no idea what I’m talking about, you should probably just skip this post). The components of the rankings are supposed to be secret, but they’re really not. In fact, one blogger – Eddie Bajek at Tigers Thoughts – believes he has pretty much cracked the formula (although he is humble enough to admit that he is likely to tag a few guys incorrectly). Eddie has shared a lot of his thought process on his blog, and I think he’ll be pretty close with most of the players.

The Elias rankings are flawed for a large number of reasons. To save time, I’ll just point out three. Ballpark effects are not taken into account. 50 homers in Coors are treated the same as 50 homers in Shea. Age/injury status are not taken into account. A 42 year old facing off-season surgery is treated the same as a 28 year old who is healthy as an ox – although the 28 year old is likely to be a better player in future years. Finally, some of the stats are just bizarre. Fielding percentage for catchers is a particular amusing stat. Catchers are awarded a putout on a strikeout – thus, catchers for teams with pitchers who have high strikeout rates get an artificial boost to their fielding percentage.

My suggestion is to scrap the entire Elias system. Instead, focus on a better measure of value. Money. If I can get 10M/yr and you can get 5M/yr, I am twice as valuable – regardless of what Elias says. Hence, my former team should receive a higher level of compensation.

How, then, do we determine type A and type B? By figuring out where their salary fits into the MLB salary structure.

Here’s one thought. Obviously, this is a rough draft, and not a final product.

Step 1: Calculate the salaries for every on MLB 25 man rosters at the end of the year. Include prorated salary bonuses and incentives that were earned.

Step 2: Determine the 90th and 80th percentile. The 90th percentile would be the cutoff for type A and the 80th percentile would be the cutoff for type B. (Obviously, this could be tweaked).

Step 3: OK, this is the hard part. What, exactly, do we measure? Total value of the contract? Average salary? Obviously, we want to avoid having teams game the system. I’ll take a stab at it.

NOTE: This refers to the player’s NEW contract, not the expiring contract.

A: Only begin this process if the team loses the player submits a request for compensation. There is not point in going through this process for a 40 year old utility infielder.

B: Determine which incentives are likely to be achieved, and add these to the base salary. I believe the NFL does this for salary cap purposes, so I believe that MLB should be able to handle this.

C: Determine how many “significant” years exist within the contract. For example, let’s assume a contract is structure this way:

Year 1: 20M
Year 2: 20M
Year 3: 20M
Year 4: 500K

Clearly, year 4 is not a significant year within the contract. Let’s set the cutoff as 70% of the highest salary. This, if the highest salary that is expected to be earned under the salary is 10M, only years with salaries of 7M or more would be counted.

[Note: this step probably seems strange. It is just in place to avoid having teams add empty years to the contract to spread the signing bonus over more years (see step D)]

D: Prorate the signing bonus over the significant years. If the contract is 10 actual years (5 significant years) and has a 50M signing bonus, we would prorate the 50M over the 5 significant years – adding 10M to each salary. We would have to determine how to handle option years. My initial thought would be to simply not include them at all.

E: Look at the salaries we end up with after jumping through these hoops. Is the highest salary at the 90th percentile? Then the player is a type A free agent. If the highest salary is at the 80th percentile, the player is a type B free agent.

F: Have MLB’s arbitrator settle any disputes.