Stats & Analysis

BigBadRob83

In the lead-up to baseball season, and doing all my fantasy prep, came across something that I had meant to do at the end of last season, but forgotten about:

Pythagorean Record
Borrowed from baseball, this stat measures how many games a team should win based on their points scored and points conceded.



So have highlighted the 'lucky' teams in green, 'slightly lucky' in yellow and 'unlucky' in red.
Probably a bit early to be doing a similar thing for season 2, but thought it was interesting anyway. The idea is that wins/losses where game is decided by one score or less is not something that team's have control over in the long run (can be very dependant on the clock, one big play, etc.)

EDIT: Formula is (PF^2.37)/(PF^2.37+PA^2.37) = winning percentage [multiplied by 16 gives expected wins]

Analysis (of sorts)
Think it definitely shows something in relation to Shane's Cardinals who were 'unlucky' last season to only win 10 games, he's already 9-0 this season.
Similarly with Adrian's 49ers, now 7-2.
It would also expect some regression from the Giants and Rams, which may or may not be happening, we'll know better at season's end.

helimachoptor

Not sure how you can call Berty lucky.. he's a beast!

BigBadRob83

helimachoptor wrote: »

Not sure how you can call Berty lucky.. he's a beast!

Haha, definitely not doing that, especially as playing him tonight.
Just putting the stats out there. I don't think any stats would predict a team to go 16-0 unless they somehow shutout every opponent

Berty44

Thats the accountant in ya coming out Rob :pac:.

Facinating analysis.

Raoul

Rob, they should really take your ideas into consideration on the LM site. I am a stat geek, love this sort of stuff

helimachoptor

Definitely Raoul, there really is a whole lot you can do with the stats!

BigBadRob83

Raoul wrote: »

Rob, they should really take your ideas into consideration on the LM site. I am a stat geek, love this sort of stuff

Yeah the main part of the LM site is the fact that it brings all the stats together, just a pity that the numbers they use aren't great.

Got good feedback from the suggestions I made on the LM forum, so fingers crossed in the next update they add in some.

Shane.C

Stats like that are very interesting. However, last season i did sim alot of my games, i didn't even play adrian last season, he played my CPU twice.

empacher

Know whose going to go moneyball on it and create a team that doesn't make big plays at all, but consistently makes 3.5 yards a play, whose defense only gives up 3 yards a play.

McG

empacher wrote: »

consistently makes 3.5 yards a play

Steve's Buccs were like that against me last gameweek. Crazy 3rd down %. Not many big plays, just really long drives made of heaps of little plays.

Shane.C

A question on the 'Power Rankings' on the LM site, it says i am top but with the Jets having much more positive stats than i do!

Why is that?

adrian522

I don't know exactly how it works but the formula is on the site:

Formula for Calculating Power Rank

((((sqr(((Points For - Points Against) + ( Strength of Schedule * 2000)) * True Strength of Schedule) + ((Wins - Losses) * 4)) * (Redzone Scoring % / 4) ) + Turnover Differential) - (Losses * 20)) + 500

Points For: Points the team has scored in total
Points Against: Points the team has given up
Strength of Schedule: The strength of the teams schedule of teams they play
True Strength of Schedule: The strength of the teams schedule in combination of the strength of each opponents schedule they play ((2 * Opponents Record) + (Opponents' Opponents Record)) / 3
Wins: Number of wins a team has
Losses: Number of losses a team has
Turnover Differential: The difference in take-aways vs. turnovers
Redzone Scoring %: The percent of times a team scores a TD or FG in the redzone
sqr: Square Root
+: Addition
-: Subtraction
*: Multiplication
/: Division

McG

Shane.C wrote: »

A question on the 'Power Rankings' on the LM site, it says i am top but with the Jets having much more positive stats than i do!

Why is that?

mainly down to you having a tougher schedule to this point which isn't a surprise due to the relative strength of the NFC over the AFC this season.

BigBadRob83

Thought this might be useful for trade season, as a prelude to my Player Progressions Analysis.

List of all players on all rosters to make comparisons easier.

BigBadRob83

Did same for Season 1, now updated for Season 2.
Borrowed from baseball, this stat measures how many games a team should win based on their points scored and points conceded.

Formula is (PF^2.37)/(PF^2.37+PA^2.37) = winning percentage [multiplied by 16 gives expected wins]



So have highlighted the 'lucky' teams in green, 'slightly lucky' in yellow and 'unlucky' in red.
The idea is that wins/losses where game is decided by one score or less is not something that team's have control over in the long run (can be very dependant on the clock, one big play, etc.). However, can be slightly misleading for teams that suffered(or inflicted) a blowout or two.
As an example, the Rams looked to be very lucky, but this is unlikely to be the case as they suffered a few blowout losses.

Shows just how good Berty was, that expected wins are 15.4!
NFC North is interesting with Bears and Lions showing big differences between actual record (Bears 10, Lions 9) and expected (Lions, 10.2, Bears 8.5)

helimachoptor

Lost 3 games by 1 score or less and won 2 games by 1 score or less so it's farily accurate.
Nice one Rob

Pudsy33

Wow if that was my lucky season I should pack it in now :pac: