Average Future Overall for each drafted position

[finkellll]

So I did some draft analysis number crunching for GML, and the only thing that is general enough for the community at large is this graph here, which shows the average future overall for each drafted position (1.1 is 1, 2.1 is 33, etc) based on my own scouts in GML, taken after the pre2 sim of that player's rookie year. There are 18-21 players for each position, so I would expect another 10 years would tighten up the graph, but not really change the overall look.


If you were a GML owner between 2115-2137 and want to see how you stack up against the other owners in that timeframe, the full thread is here: 2 decades of draft analysis - High Effort Edition (http://gml.fof-belco.com/forum/showthread.php?t=6756)


https://i.postimg.cc/T3TCTNdx/average-f-ovr.png

[Ushikawa]

Seems about right there are usually 20 true difference makers then 30 value-add starters and another 50 solid starters. That's actually how I use DA to tier out the list.

[MIJB#19]

Have you considered looking at the start of regular season 3 (more or less when most unmasking will be worn off) of these players and likewise focus on their "current ability"?

[finkellll]

I have post-year 5 data for SP but nothing for multi-player. Collecting it in GML wouldn't be useful outside of GML because it is a 0 injury league with the entire regular season simmed at once. Because there are no injuries it is tough to give the rookies (especially OL) the reps they need to develop, and because the entire season is simmed in one shot there aren't any games where the backups are played because its an obvious win/loss this week. The year 3 development data would be severely bleak compared to a more normal pace MP league.

The advantage of GML is that we draft a lot. Every other week is a new draft. It allows for relatively fast data collection in the MP scene, and it also requires effort to squirrel away player_personnel.csv files each year because it gets overwritten each data export.