I’ve been having a look at all the performance data from all meaningful weeks of the DynaBowl season for all players on the main rosters (so not including IR or taxi squads).

It’s not really that meaningful to look at the total points of players not started as some of those will be classified as out, while others will specifically be back-ups (see Jimmy Garropolo) so wouldn’t be expected to score anything.

PLOB isn’t looked at, but is examined on the MFL website in the Power Ranking data. There might be something that can be done with the raw data and I’ll hand that over to anyone else who wants to have a go.

I’m looking at scoring by position groups, consistency of scoring, and high scoring performances that were left on the bench (many of which are just unlucky because they come from players you wouldn’t think of starting – not all of them though). There may be a few other bits and pieces that get included as I follow some loose ends…

So, to start off, a couple of tables of data that have kind of been covered in the awards. All of these are sorted by the full season total points. First up, total points for all positions:

Next up, the same again, but for offense only:

Now defence:

And finally special teams (punters and kickers only):

I’ve not really got anything to say about that. I suppose the key thing is how Neil pulled out all the stops in the playoffs, destroying everything in his path.

What I find interesting, though, is the amount of variation from one week to the next. I mean, we know there’s a fair amount of variation in fantasy football, but only 3 teams had less than 100 points between their max and min scores during the 16 weeks. That seems massive.

So how do we rate consistency? I’m going to bastardise a method of statistical evaluation to try to provide a suitable indicator. What I’ve done is look at each team’s weekly score and take the difference between that and their average score. This number is then squared, which means that the higher the difference from the average, the number generated is exponentially higher.

Make sense? No? OK, an example – in week one DynaForOne scored 125.265 points. The season average was 194.133 so the difference is 68.868. 68.868 squared is 4742.78. That is a high consistency score because the weekly score was a long way from the average score.

Of course, consistency is not always bad. The same calculation for week 16 shows DynaForOne (DFO) have a consistency score of 8812.93. That week DFO scored the highest score of any team in any week of the season.

Funnily enough, these two performance, week 1 and week 16 for DFO, are the 1st and 6th least consistent performances of the season (ie with the greatest variation from the team’s weekly average). DFO also has the 7th least consistent score (week 4, another low score).

What I have done to rank the consistency is take the average of all the weeks consistency ratings and square root them. This gives us a number which probably has little meaning beyond “the lower the number the more consistent the team was over the season”.

So what’s the value in consistency? I have no idea. The top two most consistent teams are the least successful and the third most successful (by total points) so it’s difficult to say you do or don’t want to be consistent. The least consistent team won the league.

However, let’s look at DFO more closely. After week 4, DFO only scored below 190 points once. Those first 4 weeks really seemed to come from a different team. Let’s also ignore week 16, because that massive haul of 288 points was a real outlier. So, for weeks 5-15, how consistent was DFO?

For those 11 weeks, this is what those stats look like:

From least consistent to most.

I think the main thing is that if you score a lot of points, it’s really handy if you can do it consistently. Not that it helped Dyna Hard, of course.

So, given we’ve established that it’s pretty meaningless, it’s time to look at it in a bit more depth…

Offensive consistency!

What I find interesting is that Dynasore Losers and Dyna Hard were 2nd and 3rd in overall consistency but are 9th and 10th in offensive consistency. We’ll look at defensive consistency in a moment, but this suggests (to me) that they will be low there too, and that the overall consistency may be down to chance (mismatching high and low offensive and defensive scoring).

As for DFO… for those 11 weeks, there again jumped to the top of the consistency rankings, though it’s not quite as clear cut…

Now to defense!

I was wrong! Dynasore Losers had the second most consistent defence. Well done them. Dyna Hard were mid-table.

And DFO? Again, for those 11 weeks, they shot to the top of the charts…

What’s that? Special teams? You want to look at special teams consistency? OK. It’s widely thought of as the most random of all areas, though because it’s only 2 players and not much scoring, the numbers should all be much lower. Let’s see what those numbers say…

I’m not looking at DFO for that 11 week period. It’s meaningless. MEANINGLESS I tell you.

Still, at least the Bombermen finished at the top of one of these…

Now, I promised you a look at scoring by position and high scoring players on the bench… those will follow in a a future post (or posts). This is enough to start with, isn’t it?