Announcement

**Rawrster** · 07-22-2016, 06:58 PM

Great podcast! There is definitely some good information here and it is nice to compare how I use the tool with others. It is also good to see I'm not the only one who doesn't like drafting at 39

I did find it interesting that, for as much as you talked about (correctly) comparing Bars and Combine, you never mentioned the 'Rate' field. I probably use that field more than any other. I can't tell you exactly how the number there is generated, but it seems to give me a really good balance between Combine and Bars. I can sort by that and know that, usually, the guys at the top of my list have both solid combines and bars. I find this particularly helpful in the mid-rounds when I need to sort through a large number of 'average' players.

**Red Zone** · 07-22-2016, 07:36 PM

I used to think the rate column had something to do with overall rating, but ive seen too many guys drafted that had the negative combine totals and low rate columns...and they turn out ok...so Im not sure what rate relates to exactly...and I have been paying less attention to it. When I find it most helpful is in the early rounds when 2 players have similar combine or bars totals...I guess there are many ways to use the DA

Like I said on the show, If someone wants to explain to everyone what to use as a guide - like comparing bars and combines - at a 39 setting so that it isnt a guessing game...then i will stop saying 39 is a guessing game. So far no one has been able to do that

**Rawrster** · 07-22-2016, 10:22 PM

Has anyone tried just asking gstelmack for additional clarification of the tool? It seems to have a lot of unknowns for something developed by an active member of the FoF community.

As for Standard Deviation.... it is showing how many "steps" a players is from being Average at his position. How big is a "step"? Well... See how far a player is from Average. Square that number. Now do that for all the players. Take the average of all those numbers. Now take the square root of that number. That's the size of your step. At least in theory....

I did the math with the Sole scores of all the QBs in a recent draft I did and the math is ever so slightly off for some reason. So the code is handling Standard Deviation in a slightly different way than I am expecting. Steps seem to be consistent from the Median instead of the Average, which is throwing me off a tad. I'm not sure why he is using Median (middle of the pack player, not average) instead of Average, but I'm also not an expert in Standard Deviations and there are some complicated looking variants when I looked it up.

So... I guess it is more accurate to say that, in this case, it is measuring how many "steps" a player is away from the middle player in the group. This deviation should look about like a bell curve so you can expect about 2/3 of the players to be within 1 "step" of the median player (1/3 above that player, 1/3 below). The majority of the remaining players should be within one to two "steps" of the median player. A special (or sucky) few players will be greater than two "steps" away.

So, if you are looking at forty yard dash times for the RBs in the draft and a player has a standard deviation of 0.75, you know he is faster than most other RBs in the draft. If another RB has a standard deviation between 1.5-2.0 he is leaving all other RBs in the dust. These numbers are only relative to THIS DRAFT CLASS though. RBs with a 4.40 forty time could have a standard deviation of 0.6 in one draft class and 0.75 in another draft class because most the RBs in that second class are slower.

So, you can look at those Standard Deviation numbers to see how players stack up against one another, but.... the colored boxes around that player's combine results are already helping do that for you. You can look to see a player has a standard deviation of 1.58 or you could just see he has a bright green box around his forty time.

**Rawrster** · 07-22-2016, 11:08 PM

StandardDev.jpg

An example of how you could (fairly simply) use Standard Deviation is as follows.

Looking at 40 yard dash times, Leroy Brummit has a Standard Deviation of 0. That means he has middle of the road speed for a RB in this draft at 4.67 seconds. Andres Stuart is much faster (as seen by the bright green box) with a speed of 4.52 seconds and a deviation of 1.61. The difference in speed between Brummit and Stuart is 0.15 seconds. 0.15 = 1.61X, so X=.093= a Standard Deviation for a RB forty time. This tells us that about two-thirds of all RBs in this draft have a 40 time between 4.577 and 4.763 seconds. It also means that Stuart is a pretty fast RB (green box already told us that).

I just went through and counted up the running backs in the draft with forty times between 4.58 and 4.76 seconds. 33 out of 55 (60%) of backs who ran the forty fell within that range. Not quite the 66% we were expecting, but we are fairly close.

Announcement

Draft Analyzer Discussion on FOF radio

Draft Analyzer Discussion on FOF radio

Comment

Comment

Comment

Comment