![[DEV] ISFL Forums](http://dev.sim-football.com/forums/images/emerald/logo.png "[DEV] ISFL Forums"){kind=logo}
+- [DEV] ISFL Forums (http://dev.sim-football.com/forums)
+-- Forum: Community (http://dev.sim-football.com/forums/forumdisplay.php?fid=5)
+--- Forum: Media (http://dev.sim-football.com/forums/forumdisplay.php?fid=37)
+---- Forum: Graded Articles (http://dev.sim-football.com/forums/forumdisplay.php?fid=38)
+---- Thread: * NEW STATISTIC: TPE Effectiveness Ratio (/showthread.php?tid=3034)
* NEW STATISTIC: TPE Effectiveness Ratio - cosbornballboy - 08-07-2017
This new statistic, TPE Effectiveness Ratio, is measured differently for each positition. Specifically, here is how it is measured for each position:
QB - TPE : Passer Rating
HB - TPE : Yards Per Carry * Rushing Touchdowns
WR & TE - TPE : Yards per Reception * Receiving Touchdowns
DL - TPE: Sacks per games * Tackles For Loss per game
LB & DB - TPE: Tackles per game * Forced Fumbles per game * Interceptions per game
It's sort of a complex system, but it, at a very general level, determines how effective TPE is at garnering key statistics at each position.
So now, let's check out the Top 5 leaders in Passer Rating for S2 through Week 4:
1. Ethan Hunt - SJS - 92.3 PSR
2. Chris Orosz - YKW - 83.7 PSR
3. Scrub Kyubee - BAL - 77 PSR
4. King Bronko - ARI - 69.8 PSR
5. Jameis Christ - PHI - 69.3 PSR
Now, we look at the Top 5 in TER (TPE Effectiveness Ratio) You want this to be as low as possible.
1. Ethan Hunt - SJS - 1.68 TER
2. Jameis Christ - PHI - 2.40 TER
3. Scrub Kyubee - BAL - 2.71 TER
4. Chris Orosz - YKW - 3.37 TER
5. King Bronko - ARI - 3.38 TER
What we see here is that in QBs, TPE is more important to PSR than the players you have around you, seeing that QBs from Philly and Baltimore are near the top, even when those teams are near the bottom of their divisions.
Next, let's look at the top 5 Running Backs in TER (min. 50 ATT) here, if a RB doesn't have a TD, his TER is incalcuable
1. Darlane Farlane - BAL - 21.11 TER
2. Jack Durden - SJS - 22.68 TER
3. Boss Tweed - COL - 25.38 TER
4. Omar Wright - PHI - 60.79 TER
5. Ardie Savea - LVL - 62.73 TER
Here, a lot of it doesn't matter until the end of the season because if a RB has 0 or 1 TD, it skews their results a bit.
This was just a little bit of a preview into the system, and I will be calculating at the end of the season.
Graded
* NEW STATISTIC: TPE Effectiveness Ratio - Bzerkap - 08-07-2017
If you're just multiplying by TDs then why not start at 1? That way it's just an evaluation of yards per carry unless they get that TD bonus.
Honestly though, I would prefer the TDs just be a straight addition instead of a multiplying factor.
Something like this:
TPE : YPC + (TDs * 0.5)
The way it is right now values short yardage backs with tons of TDs rather than drive sustainers with tons of yards.
* NEW STATISTIC: TPE Effectiveness Ratio - Perry87 - 08-07-2017
Wait so a LB with 30 tackles, 12 FFs but 0 INTs gets a score of 0?
* NEW STATISTIC: TPE Effectiveness Ratio - cosbornballboy - 08-07-2017
(08-07-2017, 12:09 PM)Bzerkap Wrote:If you're just multiplying by TDs then why not start at 1? That way it's just an evaluation of yards per carry unless they get that TD bonus.
Honestly though, I would prefer the TDs just be a straight addition instead of a multiplying factor.
Something like this:
TPE : YPC + (TDs * 0.5)
The way it is right now values short yardage backs with tons of TDs rather than drive sustainers with tons of yards.
I'll use this one at the end of the year that's a good idea
* NEW STATISTIC: TPE Effectiveness Ratio - Bzerkap - 08-07-2017
(08-07-2017, 02:14 PM)cosbornballboy Wrote:I'll use this one at the end of the year that's a good ideaFor sure. Talk to @timeconsumer, @7hawk77, and @kckolbe about some stat stuff you could do. I know I'm missing some people but those are the guys I've seen do some prominent statistics work, here on the site or elsewhere.
* NEW STATISTIC: TPE Effectiveness Ratio - cosbornballboy - 08-07-2017
(08-07-2017, 12:13 PM)Perry87 Wrote:Wait so a LB with 30 tackles, 12 FFs but 0 INTs gets a score of 0?
I'm still trying to work out the details. I might make it times INT (if applicable)
* NEW STATISTIC: TPE Effectiveness Ratio - 7hawk77 - 08-07-2017
(08-07-2017, 02:15 PM)Bzerkap Wrote:For sure. Talk to @timeconsumer, @7hawk77, and @kckolbe about some stat stuff you could do. I know I'm missing some people but those are the guys I've seen do some prominent statistics work, here on the site or elsewhere.
I honestly don't know the right or wrong way to do this, but if I were to do this, I'd probably do it something like this:
First the data...
Total TPE count:
Logan Noble- COL (337 TPE)
Mike Boss- OCO (302 TPE)
Chris Orosz- YKW (282 TPE)
King Bronko- AZ (236 TPE)
Josh Bercovici- LVL (224 TPE)
Scrub Kyubee- BAL (209 TPE)
Jameis Christ- PHI (166 TPE)
Ethan Hunt- SJS (155 TPE)
Shawn Brady- PHI (107 TPE)
Vincent Draxel- AZ (107 TPE)
Next I'd probably create a TPE rating for each player. Formula below:
QB TPE rating = QB total TPE/(highest TPE of any QB)
Logan Noble 100% which is just 337/337
Mike Boss 89.6% which is 302/337
Chris Orosz 83.7%
King Bronko 70%
Josh Bercovici 66.5%
Scrub Kyubee 62%
Jameis Christ 49%
Ethan Hunt 46%
Then Compare them to the Passer Rating for S2:
1. Ethan Hunt - SJS - 92.3% PSR
2. Chris Orosz - YKW - 83.7% PSR
3. Scrub Kyubee - BAL - 77% PSR
Then something like Passer Rating minus TPE rating = TPE effectiveness
TPE Effectiveness means they are playing that much percentage better than they should with the TPE that they have earned
So for example:
Ethan Hunt: 92.3% - 46% = 46.3%
So Ethan Hunt plays 46.3% better than he should for the TPE that he has.
Chris Orosz 83.7% - 83.7% = 0
Chris plays standard exactly equal to how he should given the TPE he has.
Scrub Kyubee 77% - 62% = 15%
Scrub plays 15% better than he should for the TPE that he has.
If the outcome is negative, it means they are playing worse than they should for the TPE that the player has.
etc...
Cheers
* NEW STATISTIC: TPE Effectiveness Ratio - timeconsumer - 08-07-2017
If you're writing stats pieces with fewer than 6 beers in your system, you're doing it wrong.
That's all I have to add at the moment.
* NEW STATISTIC: TPE Effectiveness Ratio - Bzerkap - 08-07-2017
(08-07-2017, 02:30 PM)7hawk77 Wrote:I honestly don't know the right or wrong way to do this, but if I were to do this, I'd probably do it something like this:That's a good thought on it. I like that as a jumping off point
First the data...
Total TPE count:
Logan Noble- COL (337 TPE)
Mike Boss- OCO (302 TPE)
Chris Orosz- YKW (282 TPE)
King Bronko- AZ (236 TPE)
Josh Bercovici- LVL (224 TPE)
Scrub Kyubee- BAL (209 TPE)
Jameis Christ- PHI (166 TPE)
Ethan Hunt- SJS (155 TPE)
Shawn Brady- PHI (107 TPE)
Vincent Draxel- AZ (107 TPE)
Next I'd probably create a TPE rating for each player. Formula below:
QB TPE rating = QB total TPE/(highest TPE of any QB)
Logan Noble 100% which is just 337/337
Mike Boss 89.6% which is 302/337
Chris Orosz 83.7%
King Bronko 70%
Josh Bercovici 66.5%
Scrub Kyubee 62%
Jameis Christ 49%
Ethan Hunt 46%
Then Compare them to the Passer Rating for S2:
1. Ethan Hunt - SJS - 92.3% PSR
2. Chris Orosz - YKW - 83.7% PSR
3. Scrub Kyubee - BAL - 77% PSR
Then something like Passer Rating minus TPE rating = TPE effectiveness
TPE Effectiveness means they are playing that much percentage better than they should with the TPE that they have earned
So for example:
Ethan Hunt: 92.3% - 46% = 46.3%
So Ethan Hunt plays 46.3% better than he should for the TPE that he has.
Chris Orosz 83.7% - 83.7% = 0
Chris plays standard exactly equal to how he should given the TPE he has.
Scrub Kyubee 77% - 62% = 15%
Scrub plays 15% better than he should for the TPE that he has.
If the outcome is negative, it means they are playing worse than they should for the TPE that the player has.
etc...
Cheers
* NEW STATISTIC: TPE Effectiveness Ratio - 7hawk77 - 08-07-2017
Yeah, there is probably a more accurate way to do this. But I think this is pretty fair.
It's almost assumed that the player with the most TPE should have the best stats. So that's why I compared every other QB to Noble to create the baseline TPE ratio.
Also it's pretty interesting that I got a 0% from Chris Orosz using this method. Always feels good to get even numbers like that.
The big kicker is the passer rating gets effected by so many other things...
1.) Home field advantage.
2.) Did the QB play against strong defenses?
3.) How is your offensive line?
4.) How is your offensive weapons?
and probably more factors that I can't think of.
I think the sabercats have the strongest OL currently and some beast receivers. On top of that our run game isn't exactly slouching so it gives Hunt a ton of options and breathing room. I think this probably has to do with his performance vs TPE vs other QBs.
(I'm not saying he is bad, I think he's used his TPE well, I just don't see how he could use his TPE to be 46% more effective compared to other QBs.)