04-01-2023, 03:34 AM
(This post was last modified: 04-01-2023, 08:54 PM by Caleb_H. Edited 1 time in total.)
First of all, thank you to the good people in Discord, who have helped walk me through how to get some of the data I’m about to give to you. Without them, I’d be directionless, frankly.
People seemed to like my perspective in the first article I wrote about the passing stat history of the ISFL (which I will expand on in another article at a later date), so I’m going to take a look at another position: wide receiver. Instead of a historical look, though, I’m going to focus on this season. I understand receiver has more users than any other offensive position in the ISFL, so, hopefully, some people are interested in analysis.
I wanted to create one holistic stat that I could look at, if I wanted to vote on All-Pro or Pro Bowl or whatever we have on here to evaluate the receiver position. This is kind of how my process worked:
1) Yards are good, but they don’t include touchdowns.
2) Including touchdowns is good, but it doesn’t account for how many times a player is thrown the ball (ie: Was the player efficient with the ball or did he simply get force-fed?)
3) Even including targets into the mix, that doesn’t account for penalties, penalty yards and fumbles, which shockingly do matter a decent amount in this sim if you’re going to rank single seasons.
What I ended up settling on was adjusted yards per target value, which sounds like a mouthful. Here’s what it means:
Adjusted yards uses roughly the same formula as AY/A does in the passing game, but we’re going to include penalties and penalty yards in the mix. So a touchdown would count as a bonus of 20 yards and penalty yards would be subtracted from the adjusted yards total. In AY/A, an interception comes with a -45 yard penalty, which I figure we can split in half for fumbles (due to fumble luck.)
That gives us adjusted yards. From there, we can divide that total number by the number of targets and penalties a player received over the season to give us adjusted yards per target. This only works under the assumption that a player’s penalty and fumble came on the offensive side of the ball, which I’m uncertain of at the moment. I simply do not know one way or the other. I could be wrong here and I'm very aware of it!
The next step is the value step, when we take a player’s adjusted yards per target (AY/target) number and compare it to the league average at the position (min. 50 targets) to create a +/- yards metric. We do this by subtracting the league average AY/target number from an individual’s AY/target number and multiplying it by the number of targets and penalties a player received.
Above is what that looks like. I know it can look very noisy, but just know that lime green is the top 20 percentile in the league, bright red is the bottom 20 percentile in the league and each color is its own 20 percentile bucket. It's ranked by adjusted yards per target value, on the far right, but includes all the raw stats and other steps along the way to get to the final number.
So as you can see, the top receiver in the league -- by this calculation -- was worth about 313 adjusted yards above the average of the league. Congrats, Arizona’s Captain Billy Stinkwater on both the ring and the crown. If you compare Stinkwater’s advanced numbers to Yellowknife’s Calvin Spiff’s (not picking on you!) you can see how wide that gap can be. Spliff had 722 adjusted yards on 142 plays (5.4 yards per) while Stinkwater’s 116 were good for 1,268 adjusted yards (10.8.) Pretty significant!
Obviously, RNG and a quarterback play probably come into play here, but it’s worth noting that Honolulu had both the fourth-ranked receiver (Leek Mai-Heinous, great name) and second-to-last-ranked receiver (Rockbot Rockbo) using this metric.
Here are some oddities I noticed, over multiple of these metrics:
Cape Town’s Derrick Prince easily has the most unique statline. Yes, his fourth-ranked 1,394 raw receiving yards is high, but he also racked up 10 penalties for 85 penalty yards and five fumbles, which brought down his adjusted yards per target to 17th in the league. The same is close to true with Eeeeeeee phenssta (third-ranked in receiving yards with six penalties and four fumbles.)
To take a look at what’s going on there, I decided to look at players’ relevant stats, including TPE, to see if how they built their players had anything to do with these results.
The table above shows the raw stats, advanced stats and the player stats (is that the word?) for this season’s ISFL crop.
Interestingly enough, if you sort by raw yards, phenssta and Prince are the only two receivers who rank in the top-17 in receiving yards who have below 70 in intelligence or in phenssta’s case have lower than a competitiveness stat of 63.
Generally, adjusted yards per target value does track with TPE, but there are exceptions. Let’s highlight some above and underperformers.
The one player who greatly outplayed his TPE is Sarasota’s Johnny Blaze Jr., the only player to rank in the bottom 40 percent of TPE and in the top 40 percent in adjusted yards per target. The remaining four players we’re highlighting are all receivers who underperformed, relative to their TPE. New York’s Mandrews McHollywood, Cape Town’s Erwin Kemp, Orange County’s Thomas Robinson and Yellowknife’s Calvin Spiff all were in the top 40 percent of TPE (McHollywood and Robinson in the top 20 percent) but ranked in the bottom 40 percent in adjusted yards per target value (Spliff last in the league.)
For whatever it’s worth, it does seem like speed receiver is the meta in the game. This past season, among qualifiers, the speed receivers combined for 1,292 adjusted yards above the league average. That was over 18 receivers (an average of 72 yards above average per player.) Slot receivers, the next biggest group, were down 1,015 yards versus the average over 11 receivers (92 yards below the average per player.) Possession receiver was the only other significant group, down 549 yards below average over six players (92 yards below the average per player.) The only return specialist was Austin Morley, who was 273 yards above the average.
Above is a chart of the “off-meta” or non-speed receiver archetype receivers from last season. It’s worth noting that the players outside of speed receiver did end up doing solid as long as they hit 90 in hands and 95 in speed, but that there are a lot more low TPE receivers in this group, from a percentage standpoint. There are some disappointing performances relative to TPE, though, like Kemp, Robinson, Rockbo and Spiff.
--
I hope none of this is taken in the wrong way. I’m trying to understand this game and figured I’d share some of my findings!
These numbers would be even better if we knew how many routes players ran, but as far as I know, that data doesn't exist. Let me know if I'm wrong. Also, I'd like to hear what else you think I should take a look at. I would like these posts to be discussions, if possible.
No, I am not someone's multi.
People seemed to like my perspective in the first article I wrote about the passing stat history of the ISFL (which I will expand on in another article at a later date), so I’m going to take a look at another position: wide receiver. Instead of a historical look, though, I’m going to focus on this season. I understand receiver has more users than any other offensive position in the ISFL, so, hopefully, some people are interested in analysis.
I wanted to create one holistic stat that I could look at, if I wanted to vote on All-Pro or Pro Bowl or whatever we have on here to evaluate the receiver position. This is kind of how my process worked:
1) Yards are good, but they don’t include touchdowns.
2) Including touchdowns is good, but it doesn’t account for how many times a player is thrown the ball (ie: Was the player efficient with the ball or did he simply get force-fed?)
3) Even including targets into the mix, that doesn’t account for penalties, penalty yards and fumbles, which shockingly do matter a decent amount in this sim if you’re going to rank single seasons.
What I ended up settling on was adjusted yards per target value, which sounds like a mouthful. Here’s what it means:
Adjusted yards uses roughly the same formula as AY/A does in the passing game, but we’re going to include penalties and penalty yards in the mix. So a touchdown would count as a bonus of 20 yards and penalty yards would be subtracted from the adjusted yards total. In AY/A, an interception comes with a -45 yard penalty, which I figure we can split in half for fumbles (due to fumble luck.)
That gives us adjusted yards. From there, we can divide that total number by the number of targets and penalties a player received over the season to give us adjusted yards per target. This only works under the assumption that a player’s penalty and fumble came on the offensive side of the ball, which I’m uncertain of at the moment. I simply do not know one way or the other. I could be wrong here and I'm very aware of it!
The next step is the value step, when we take a player’s adjusted yards per target (AY/target) number and compare it to the league average at the position (min. 50 targets) to create a +/- yards metric. We do this by subtracting the league average AY/target number from an individual’s AY/target number and multiplying it by the number of targets and penalties a player received.
Above is what that looks like. I know it can look very noisy, but just know that lime green is the top 20 percentile in the league, bright red is the bottom 20 percentile in the league and each color is its own 20 percentile bucket. It's ranked by adjusted yards per target value, on the far right, but includes all the raw stats and other steps along the way to get to the final number.
So as you can see, the top receiver in the league -- by this calculation -- was worth about 313 adjusted yards above the average of the league. Congrats, Arizona’s Captain Billy Stinkwater on both the ring and the crown. If you compare Stinkwater’s advanced numbers to Yellowknife’s Calvin Spiff’s (not picking on you!) you can see how wide that gap can be. Spliff had 722 adjusted yards on 142 plays (5.4 yards per) while Stinkwater’s 116 were good for 1,268 adjusted yards (10.8.) Pretty significant!
Obviously, RNG and a quarterback play probably come into play here, but it’s worth noting that Honolulu had both the fourth-ranked receiver (Leek Mai-Heinous, great name) and second-to-last-ranked receiver (Rockbot Rockbo) using this metric.
Here are some oddities I noticed, over multiple of these metrics:
Cape Town’s Derrick Prince easily has the most unique statline. Yes, his fourth-ranked 1,394 raw receiving yards is high, but he also racked up 10 penalties for 85 penalty yards and five fumbles, which brought down his adjusted yards per target to 17th in the league. The same is close to true with Eeeeeeee phenssta (third-ranked in receiving yards with six penalties and four fumbles.)
To take a look at what’s going on there, I decided to look at players’ relevant stats, including TPE, to see if how they built their players had anything to do with these results.
The table above shows the raw stats, advanced stats and the player stats (is that the word?) for this season’s ISFL crop.
Interestingly enough, if you sort by raw yards, phenssta and Prince are the only two receivers who rank in the top-17 in receiving yards who have below 70 in intelligence or in phenssta’s case have lower than a competitiveness stat of 63.
Generally, adjusted yards per target value does track with TPE, but there are exceptions. Let’s highlight some above and underperformers.
The one player who greatly outplayed his TPE is Sarasota’s Johnny Blaze Jr., the only player to rank in the bottom 40 percent of TPE and in the top 40 percent in adjusted yards per target. The remaining four players we’re highlighting are all receivers who underperformed, relative to their TPE. New York’s Mandrews McHollywood, Cape Town’s Erwin Kemp, Orange County’s Thomas Robinson and Yellowknife’s Calvin Spiff all were in the top 40 percent of TPE (McHollywood and Robinson in the top 20 percent) but ranked in the bottom 40 percent in adjusted yards per target value (Spliff last in the league.)
For whatever it’s worth, it does seem like speed receiver is the meta in the game. This past season, among qualifiers, the speed receivers combined for 1,292 adjusted yards above the league average. That was over 18 receivers (an average of 72 yards above average per player.) Slot receivers, the next biggest group, were down 1,015 yards versus the average over 11 receivers (92 yards below the average per player.) Possession receiver was the only other significant group, down 549 yards below average over six players (92 yards below the average per player.) The only return specialist was Austin Morley, who was 273 yards above the average.
Above is a chart of the “off-meta” or non-speed receiver archetype receivers from last season. It’s worth noting that the players outside of speed receiver did end up doing solid as long as they hit 90 in hands and 95 in speed, but that there are a lot more low TPE receivers in this group, from a percentage standpoint. There are some disappointing performances relative to TPE, though, like Kemp, Robinson, Rockbo and Spiff.
--
I hope none of this is taken in the wrong way. I’m trying to understand this game and figured I’d share some of my findings!
These numbers would be even better if we knew how many routes players ran, but as far as I know, that data doesn't exist. Let me know if I'm wrong. Also, I'd like to hear what else you think I should take a look at. I would like these posts to be discussions, if possible.
No, I am not someone's multi.