Welcome to week 11. The playoff picture is starting to shape up but I think other than Chris, every other team is technically still alive. We'll look into the playoff picture in a future week but today I'll show you a slightly different exercise I did to show whether or not a team has been unlucky / lucky.
For the most part, everyone uses PF as the measure to show the "real" rankings. If your PF is high but your wins are low, you basically chalk it up to bad luck. But what PF doesn't do well is show week to week variance. If you score 400 points one week but then 50 for then next 4, your PF will show 500, making you look like a pretty average team. But in actuality, your team sucks and should be 1-4. So if we look at week to week scoring, will it show something different? Or will it just line up with what PF shows?
The Methodology
I took everyone's points scored this season on a per week basis. Here is that information charted out (The font color was an accident but I actually like it because the trend of fill colors is more important to note than the numbers themselves).
The darker the green, the closer you are to the top score, meaning the more % you had to win that particular week. The darker the red, the closer you are to the bottom score, meaning the more % you had to lose that particular week.
Next, I found the median of the dataset, which landed us at 107.5. Basically, if you score above this you have more than a 50% chance of winning and if you score below this, you have less than a 50% chance of winning. The standard deviation of scores is about 24 points so I used the standard deviations to approximate how many wins you should get based on how you scored weekly rather than how many wins. Anyone half a standard deviation higher got .6 expected wins. If you were .5-1.5 standard deviations higher, you got .85 wins. And if you were more than 1.5 standard deviations away from the median score, you should have been guaranteed a full win. The same logic applied to losses with decreasing expected win values. So the chart looks something like this:
- I've filtered this by my "Luck" metric which takes Actual Wins - Expected Wins. Positive numbers mean you've been lucky this year while negative numbers indicate that the football gods hate you.
- Jon remains the unluckiest player this year so far. I knew he had some bad weeks so I thought that his 200 point outing was boosting his overall PF. While his expected wins standing is slightly less than his PF standings, he is clearly, without a doubt, the unluckiest player of the year. He should be 4th but he's currently sitting in 11th, looking up.
- The person who stole all of Jon's luck? David Y. He should actually be in Jon's position at 11th in both PF and EW but he's somehow in 5th taking a whopping 2.4 wins he should never have had in the first place. Sometimes it's better to be lucky than good.
- Mike has also seen a lot of luck, although it's just taken his team from good to great. He'd still be a playoff bye team even if he stuck with his expected win total.
- Simon and Rich probably have the biggest gripes after Jon and although both would be fringe playoff teams, that's better than having to fight for your life every week to get in.
- Unsurprisingly, Chris's failure has less to do with his luck than just being a really bad fantasy football manager.
- Aside from Jon, the person who has the biggest gripe is actually Jung. He should be 3 spots higher and with his top heavy lineup, he could have been in the running for a bye which would have greatly increased his chances at his first title.
- After David, Dan and Paul have been the biggest beneficiaries of scheduling luck. They should both be much lower and possibly not even in the playoff race
No comments:
Post a Comment