![[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: *Strength of Schedules (/showthread.php?tid=9124)
Pages:
1
2
*Strength of Schedules - bovovovo - 06-12-2018
At the end of the NSFL season, every team will have played every other team twice, once at home and once away.
This means that everybody's strength of schedule is kinda the same, or at least very similar. Bottom feeder teams don't get to play themselves so their strength of schedule (SOS) is a little harder, and top tier teams don't have to play themselves so their SOS is a little easier. There is a method to help minimize this effect though, which I'll get into here in a bit.
But right now, as of week 7, they aren't completely equal. So I thought it'd be cool to look into it a bit. This is absolutely not the most accurate and in-depth SOS break down method you'll see out there, I'm just bored and could slap this together relatively quickly because it's still interesting.
First, we take every team's current Win Rate
(team) (record) -> (Win Rate)
6-1 -> 0.86
5-2 -> 0.71
4-3 -> 0.57
3-4 -> 0.43
5-2 -> 0.71
3-4 -> 0.43
1-6 -> 0.14
1-6 -> 0.14Wow amazing. So now with this we take each team's schedule and look at who all they've played, and we average together the win rates of their opponents
I'm not gonna lay it all out here but for example... The Wraiths have played NOLA twice, Hawks twice, Outlaws, Liberty, and Otters. So we take each of those win rates and average them together. (0.43 + 0.14 + 0.43 + 0.57 + 0.71 + 0.57 + 0.71 / 7 = 0.51)
I put them from highest opponent's win rate to lowest
(team) (Opponent's Win Rate)
0.59 0.59 0.57 0.51 0.51 0.42 0.41 0.39We could actually go ahead and stop here if we wanted, and there's some neat observations here.
~The Yeti have had the easiest schedule so far and the Hawks and NOLA have had the hardest.
~The Wraiths, with the best record in the league, are right in the middle
NOW LETS GO DEEPER
Now that we know each team's opponent's win rates, we're going to do basically the same thing except this time we're going to find each team's opponent's opponent's win rates.
This is because strength of schedule (obviously) affects a team's record to some degree, so after we know each team's opponent's opponent's win rate we're gonna weight each team's opponent's win rates at 66% and their opponent's opponent's win rate at 33%, and combine them together to get a hypothetically truer strength of schedule measurement.
But I'm getting ahead of myself. Here's each team's opponent's opponent's win rate ordered highest to lowest.
0.54 0.53 0.53 0.51 0.49 0.47 0.45 0.45 All this really means is that the Liberty's opponents on average had a harder schedule than the Sabercat's opponents did, for example.
So now we weight them together using this equation -> ((2*opponent's win rate) + (opponent's opponent's win rate)) / 3 to get a strength of schedule that is more reflective of everybody's schedules. This method also somewhat helps minimize the effects of "great teams don't have to play themselves so their schedules are easier" thing.
Below are the adjusted strength of schedules, highest on top. I also put how many of their games were home games out of the 7 games played so far, because that's relevant with how much the sim values home field advantage.
0.55 (4/7) 0.54 (3/7) 0.53 (3/7) 0.52 (4/7) 0.50 (3/7) 0.47 (4/7) 0.45 (4/7) 0.43 (3/7)Oh hey look, the order didn't really change at all. Except this time the Hawks are now solely in 1st with the hardest strength of schedule so far. Though they have had one more home game than NOLA has, some maybe NOLA gets a bump. I'm not sure how to quantify the effect of home field advantage on strength of schedule so I'm just not going to bother lol.
TLDR: Hawks, NOLA, and SJS have had hard schedules so far, Wraiths and Outlaws have been average, and Otters and Yeti have had easy schedules so far.
(760 words)
*Strength of Schedules - iamslm22 - 06-12-2018
Tough start to the season, we knew that stretch would be tough. Should've won at least one of the YKW games, but hoping we can close out the season strong, especially given the Philly Rebuild.
*Strength of Schedules - Roly - 06-12-2018
Great article. Love the breakdown and walk through on SOS analysis.
*Strength of Schedules - Beaver - 06-12-2018
tl;dr NOLA awesome, Yeti suck
Nice.
*Strength of Schedules - speculadora - 06-12-2018
These rankings don't match my worldview so thanks for wasting my time
*Strength of Schedules - run_CMC - 06-12-2018
Cool article Bovovovovovovo!
what I'm getting from this is the Yeti are still big succ
*Strength of Schedules - Bwestfield - 06-12-2018
I feel like home field advantage needs to be factored in here as well somehow. It's definitely a factor.
*Strength of Schedules - 124715 - 06-12-2018
I got flamed for saying the Otters had had an easy schedule in the chat last night ........
Numbers don't lie! (in more ways than one haha)
*Strength of Schedules - bovovovo - 06-12-2018
(06-12-2018, 03:53 PM)Bwestfield Wrote:I feel like home field advantage needs to be factored in here as well somehow. It's definitely a factor.
It definitely does, I just couldn't figure an accurate way to incorporate it quantitatively. Maybe some way to weigh each game based on if it was home or away?
I'm not sure. Hopefully someone smarter than I could figure it out.
*Strength of Schedules - Bzerkap - 06-12-2018
Impressive mathematics performed with a keen insight into statistical analysis. Have you submitted your thesis yet?