If you play Brent, you may score 113 points and still lose. If you play Jesse, you just need to make sure you have your lineup set. However, this is a critical point in the season, with many teams sharing the same record (seven teams at 4-4), and with the first and second seed in the playoffs all but locked up with Brent and Sean, its all about just getting into the playoffs.
I was curious the other day and asked myself, 'what do I need to score to win a fantasy football game?'. And then I was wondering to myself: Yahoo projects me to score 100+, that sounds good, but it going to translate into wins? I sure hope scoring over 100 wins me games! I would like to score 170+, but I'm no Brady, nor do I have Tom Brady on my team.
First, what I did was look at all the scores (as you would have to if you really wanted to know, but before that, I collected all the scores.... if you want to know what I truly did first) and separated them into wins and losses... I then asked, what is the break down of these scores, i.e. what are the ranges of scores that translated into wins vs losses. My thesis was that if teams scored over 120 points a game, that team would probably have a 75-80% chance of winning my game.
When looking at the league holistically, we get some higher scoring hurdles than I had first expected. The true (up to this point) scoring hurdle to ensure a victory is to score over 138.46. You will win 100% of the time if you do that. I found this by simply looking at the highest score in a loss -- it was Week 3 when Scott H lost to Brent by a mere 4.58 points, as Brent posted 143.04 (yes, I double checked my math). Scoring above 138 is a tall task, as only SEVEN teams have eclipsed that mark - with just one team, Ben, to do it twice. It is a tall order when also considering that just FIVE teams have scored above 140 points. This tells us two things: 1) It is difficult to score that high 2) it doesn't happen very often. Go to Bookings.com for something less obvious.
However, there are outliers in our data. There always is. Brent and Jesse, as stated at the very beginning, are our outliers. Without their scores accounted for (I took them completely out of the data set and all with wins/losses from their match-ups with other teams) that 138.46 mark mentioned in the last paragraph is much, much lower. It would be at 108.96 or approximately 29.5 points lower than 138.46. That is a considerable amount of points. To score 138.46 you have to have your players average ~15.38 each. I don't know about you, but I am having a tough time finding a RB that can do that, let alone a kicker and defense to put up that kind of scoring. While at our new mark of 108.96, your players need to average ~12.11 points. High? Yes, but doable. All but one team has scored over 110 points.
What this means is that you don't have to ask a few of your players to do so much, to cover for the rest of the guys sorry asses, because it will never be that balanced of an attack. In this game someone always has to pick up another players weak performance.. No thanks to you, Ryan TannehillsuckieQB! Lets look at what happened on the week's those teams that eclipsed the mark of 138.46 did.
To simplify this I just looked at the top performers for WR and RB position, i.e. the highest scoring WR and also the highest scoring RB. Then I looked at the higher of the K/DEF score and always included the QB.
QB: averaged 30+ pts, with the high of 62. That 62 number is off the charts and must be excluded (Dan posted that last week as Brees just went insane on the membrane). This brings our average down to around 27. This is still a high mark and there were two cases of where the QB scored under 20.
WR: averaged 22+. The sample set was pretty close, so I didn't have to throw any numbers out. The high was 27 and low was 17.
RB: avg. 26+. There was one week where the leading RB had around 9 points (again, Dan's performance last week - he had Brees covering everyone else). Excluding that 9 pointer, you are looking at about 28 pts. But there was one week where a RB hauled in 44 points to cover for a poor QB performance (Cameron in week 7). Taking the 44 pts out you are down to ~26 pts.
K/DEF: avg.15+. There were just two weeks where teams had their kicker or defense score over 20. Taking those out reduces this number to ~13. Interesting to note here that the lower of the two still avg. ~8. Still not a terrible performance.
Do you follow this tangent at all? Basically what I'm saying is you need your core (or just one dude to go HAM) to really sow up. You are talking about a QB score of 27, a WR at 22, and RB at 26. That is a killer week man, but you gotta have some other support here, coming in the form of a kicker or defense, to cover for a more than likely shotty TE and WR2, WR3, and RB2. But those four positions would need to put up about 90 points. That is a lot of weight on a few guys to get to 138.46, while 108.96 is less than 20 points away.
Let's circle back to my original thesis, that you will win 75-80% of the time with 120 pts? What does that mean for my chances of winning? This is when it gets even more muddy. If you aren't already lost, you may be... I took the scores of wins, put them in ranges, then divided them by the number of times that range appeared. For example, looking at the range of between 110 and 120, I found seven times a team won, versus the total number of times a number between 110 and 120 were found 10 times. Therefore, 70% of the time. However, There are two sections below. One including Brent and Jesse and another that excludes their scores and the corresponding wins scores and range scores from the sample.
*Including Brent and Jesse
Looking at the ranges in about 10 point intervals from 138.46 down you get the following:
| Scores | Chance to win | Number of Wins |
| >138.46 | 100% | 8 |
| > 120 < 138.46 | 85.71% | 12 |
| > 110 < 120 | 70.00% | 7 |
| > 100 < 110 | 38.46% | 10 |
| > 90 < 100 | 41.67% | 5 |
It may look weird that you have a high chance of winning a game if you score in the range of 90 to 100, but this is due to the fact that we have had the least amount of wins in that range.
**Excluding Brent and Jesse
Removing Brent and Jesse changes things quite a bit, as the scoring threshold in significantly reduced.
| Scores | Chance to win | Number of Wins |
| >108.96 | 100% | 19 |
| > 100 < 108.96 | 35.00% | 7 |
| > 90 < 100 | 33.33% | 3 |
| > 80 < 90 | 8.00% | 4 |
Most wins (19) have come when teams have scored above the 108.96 mark.
Finally, you now know what marks you must post to obtain victory. My thesis was correct. You score 120, you have a high probability of winning.
No comments:
Post a Comment