Player Acquisition in 2013
Having gone through the trouble of grading all the players who played in FLOF this year, I decided to use the data for a couple other things. Today, I'll look at how those players came to FLOF in 2013.
First a reminder of what this is what this isn't. The metrics I'm looking at are Value Over Median (VOM) which is basically a player's scoring minus the median for a player at their position. VOM / gm takes a season long VOM value and divides it by the number of games that guy played. I also look at VOM on a game by game basis with what I call +games and -games. A +game is a game where the player scored in the top 3rd of expected scores for a player at that position and a -game is a game where the player scored in the bottom 3rd of expected scores for a player at that position.
The goal of these metrics is to measure how much a player contributed in the context of games they played in FLOF. A player who only played one game in FLOF and not well will have a low value in these metrics even if they had a good year otherwise.
What I did next was separate all players into 4 piles based on how they were originally "acquired" by a FLOF team this year: Franchised, Practice Squad, Free Agent, and Draft. Acquired goes in quotes because franchised and practice squad players were "originally" acquired in other ways in some year prior. But I just wanted to look at this year. So a guy who is Franchised and ends up playing, gets counted as Franchised. Also not that Trade is not an option. I did not want to track everything that would imply so I ignored trades. That's not too important today, but is when I look at teams. I separated Franchising from PS because there are different expectations there. Players you franchise are people you expect to help you this year. Players on your PS you may not expect anything from this year.
A chart.
The first thing I noticed here is that VOM and VOM / gm seems too high. Shouldn't it be more around 0 by definition? So I checked my data. Turns out that most positions have higher average scores than medians. Not by much, but it adds up over 1247 player games in a season. So those higher top third type scores skew these stats a little above median... 1/2 a point per player game.
I did something similar to this many years ago and came to the conclusion that Franchising was where most of the best players were found. This analysis did not really change that conclusion but it did not really throw Draft and FA under the bus completely.
The three main piles (excluding PS) contributed a similar number of players to FLOF, but FA players contributed roughly half as many games due mostly to the fact that FA players get acquired as the season progresses and have fewer opportunities to play. So generally Franchising is better than Free Agency which is better than Drafting which is better (this year) than Practice Squad -- at least for providing value this year.
Most individual teams followed this patterns as well but not all. One team got value out of their PS. A handful also got more value out of drafting then free agency.
One more league-wide chart.
The metrics by draft round. The couple of picks that occurred after the 8th round are lumped in with the 8th round.
That's not what I think you would expect without thinking too hard about this. Apparently I should trade away all my 1st round picks for 4th round picks. Well, maybe not.
Some of this could be small sample size theater but I also think that the 1st and 2nd round include a mix of players we expect should be awesome and players that are really developmental prospects. Three players in each of the first two rounds never played a game in FLOF this year. Many ended up on the PS days after being drafted.
What do you think is up with the numbers by draft rounds?
First a reminder of what this is what this isn't. The metrics I'm looking at are Value Over Median (VOM) which is basically a player's scoring minus the median for a player at their position. VOM / gm takes a season long VOM value and divides it by the number of games that guy played. I also look at VOM on a game by game basis with what I call +games and -games. A +game is a game where the player scored in the top 3rd of expected scores for a player at that position and a -game is a game where the player scored in the bottom 3rd of expected scores for a player at that position.
The goal of these metrics is to measure how much a player contributed in the context of games they played in FLOF. A player who only played one game in FLOF and not well will have a low value in these metrics even if they had a good year otherwise.
What I did next was separate all players into 4 piles based on how they were originally "acquired" by a FLOF team this year: Franchised, Practice Squad, Free Agent, and Draft. Acquired goes in quotes because franchised and practice squad players were "originally" acquired in other ways in some year prior. But I just wanted to look at this year. So a guy who is Franchised and ends up playing, gets counted as Franchised. Also not that Trade is not an option. I did not want to track everything that would imply so I ignored trades. That's not too important today, but is when I look at teams. I separated Franchising from PS because there are different expectations there. Players you franchise are people you expect to help you this year. Players on your PS you may not expect anything from this year.
A chart.
The first thing I noticed here is that VOM and VOM / gm seems too high. Shouldn't it be more around 0 by definition? So I checked my data. Turns out that most positions have higher average scores than medians. Not by much, but it adds up over 1247 player games in a season. So those higher top third type scores skew these stats a little above median... 1/2 a point per player game.
I did something similar to this many years ago and came to the conclusion that Franchising was where most of the best players were found. This analysis did not really change that conclusion but it did not really throw Draft and FA under the bus completely.
The three main piles (excluding PS) contributed a similar number of players to FLOF, but FA players contributed roughly half as many games due mostly to the fact that FA players get acquired as the season progresses and have fewer opportunities to play. So generally Franchising is better than Free Agency which is better than Drafting which is better (this year) than Practice Squad -- at least for providing value this year.
Most individual teams followed this patterns as well but not all. One team got value out of their PS. A handful also got more value out of drafting then free agency.
One more league-wide chart.
The metrics by draft round. The couple of picks that occurred after the 8th round are lumped in with the 8th round.
That's not what I think you would expect without thinking too hard about this. Apparently I should trade away all my 1st round picks for 4th round picks. Well, maybe not.
Some of this could be small sample size theater but I also think that the 1st and 2nd round include a mix of players we expect should be awesome and players that are really developmental prospects. Three players in each of the first two rounds never played a game in FLOF this year. Many ended up on the PS days after being drafted.
What do you think is up with the numbers by draft rounds?
1 Comments:
I think you're on to something with the first couple of rounds. They are a mix of player expected to start and PS players. Also, if you look back at the draft, round 2 was terrible. You have Miller, S. Jackson, V. Jackson, D. Jackson - future note, never draft a Jackson in round 2 - all coming off the board in that round. Only saving grace for round 2 appears to be Hilton. However, you start getting into rounds 4 and 5, I noticed that was the point that defense and kickers that would start most games came off the board. In round 4, you have David and D. Jackson (both top 10), Gostkowski (#1 kicker) that were both high at their position. Then in 5, you have Posluszny and 2 starting QB's sitting there (Rivers, A. Smith). So in short, it seems to be following a trend I have noticed. Protect your offensive studs, the draft for them is way up and down. However, defense and kickers is where it's at for the draft. Why I don't have much to say for someone that takes a kicker in the 3 or 4 rounds.
Post a Comment
Subscribe to Post Comments [Atom]
<< Home