Scoresman
Footballguy
A lot of us pour countless hours into projecting/ranking/assigning value to players to prep for our drafts. While this is usually a very valuable thing to do, let's face it. Projections are mostly going to be wrong because so much luck and other uncontrollable variables comes into play with them. They are also incredibly easily attainable so everyone has them. Good projections dont give you the leg up they used to.
One thing that tends to remain constant and can unlock a treasure trove of useful information is your league's drafting tendencies and taking a deep look at them. I started analyzing this a couple years ago. I'd go so far as to say this is potentially the single most important piece of research you can do. For me, this is even more valuable than mock drafts. Mock drafts are with random people (who are probably more prepared than the average drafter). I started doing my own mocks in excel with my draft predictions and they end up looking many times more similar to my league drafts than a regular mock.
The main requirement with this is that it's obviously only going to be valuable if you generally have the same group of guys/gals drafting from year to year and also if your league's draft history is available to you. Even just a year or two works.
Let me give you an example.
My main league is an auction league and it's a tough, competitive one. For me, the single toughest part of these drafts is adapting to what happens during the draft and making quick adjustments to your strategy if necessary. I realized that if I could predict what price each player was going to go for, I could essentially predict those tough scenarios where value suddenly popped up (or vice versa) forcing you to make a quick decision.
What I did was I stacked each draft in excel by position and rank along with the price. I did this going back 3 years. What I wanted to see was how much was the most expensive QB in 2021, 2022, and 2023, same for QB2, 3, 4, etc. For this step I didn't need to know player names. I only needed to know that the most expensive QB went for X amount each year, the 10th most expensive RB went for X amount each year, etc. I noticed a striking amount of similarities each year by position and rank.
Once I did this, I started seeing some consistent patterns at each position making this year's positions fairly easy to predict. Now you can assign this year's player names by ADP or whatever and you have each player with a predicted price. Just looking at this vs. what VBD tells you the prices should be is going to be an indicator of who will end up being a value, thus giving you some predictive power in a usually very chaotic and unpredictable auction draft. For example, our league spends way less at WR than VBD says we should. This means that almost all WRs should give value in the actual auction.
You can also see things like totals spent by position each year, totals spent for starters each year, etc. Last year was the first year I went into my draft with price predictions, and it was the least stressful auction I've had in ages. I learned from this research that our league was slowly spending less each year on RB and more on WR/QB, so I predicted this trend to continue.
You can also manually tweak predictions based on homer picks and other things your leagues do. Lots of Niner fans in all my leagues so Niners get a bump in predicted pricing/ADP. But when you bump one group of players, you need to bump down another group to offset everything. Also some years, specific player tiers are more heavily loaded than others. For example, this year, starting QBs after the top 3-4 are all pretty much in one tier and that should be reflected in the predictions. For pricing I'll probably bump up the QB8-12 spots a bit higher than previous years. I'll probably offset this a little amongst the top 3 QBs, and also at other positions.
Now if your league does weird things like nominating kickers early or nominating a lot of late round players early, this can have an effect on the pricing and might make this more difficult. Luckily, mine doesn't.
Here is how my QB predictions went last year, You can see my estimate vs the actual player price. (Our league uses a lower cap than the standard $200.)
This works for snake drafts too. Just put in your draft pick history against that year's ADP and look at any patterns. I've learned the following about my snake leagues.
- In one of them, QBs get drafted a full round earlier than ADP on average and RBs about a half round later than ADP once you get past the top tier.
- In another, a ton of the other drafters pick their defense before they start drafting any backup players. I always knew defenses went too soon, now I know at what point in the draft the defense runs will start.
- One guy in my league went to college with Aaron Rodgers and knew him and always drafts him. Last year, the poor sap drafted him in the first round (superflex).
One thing that tends to remain constant and can unlock a treasure trove of useful information is your league's drafting tendencies and taking a deep look at them. I started analyzing this a couple years ago. I'd go so far as to say this is potentially the single most important piece of research you can do. For me, this is even more valuable than mock drafts. Mock drafts are with random people (who are probably more prepared than the average drafter). I started doing my own mocks in excel with my draft predictions and they end up looking many times more similar to my league drafts than a regular mock.
The main requirement with this is that it's obviously only going to be valuable if you generally have the same group of guys/gals drafting from year to year and also if your league's draft history is available to you. Even just a year or two works.
Let me give you an example.
My main league is an auction league and it's a tough, competitive one. For me, the single toughest part of these drafts is adapting to what happens during the draft and making quick adjustments to your strategy if necessary. I realized that if I could predict what price each player was going to go for, I could essentially predict those tough scenarios where value suddenly popped up (or vice versa) forcing you to make a quick decision.
What I did was I stacked each draft in excel by position and rank along with the price. I did this going back 3 years. What I wanted to see was how much was the most expensive QB in 2021, 2022, and 2023, same for QB2, 3, 4, etc. For this step I didn't need to know player names. I only needed to know that the most expensive QB went for X amount each year, the 10th most expensive RB went for X amount each year, etc. I noticed a striking amount of similarities each year by position and rank.
Once I did this, I started seeing some consistent patterns at each position making this year's positions fairly easy to predict. Now you can assign this year's player names by ADP or whatever and you have each player with a predicted price. Just looking at this vs. what VBD tells you the prices should be is going to be an indicator of who will end up being a value, thus giving you some predictive power in a usually very chaotic and unpredictable auction draft. For example, our league spends way less at WR than VBD says we should. This means that almost all WRs should give value in the actual auction.
You can also see things like totals spent by position each year, totals spent for starters each year, etc. Last year was the first year I went into my draft with price predictions, and it was the least stressful auction I've had in ages. I learned from this research that our league was slowly spending less each year on RB and more on WR/QB, so I predicted this trend to continue.
You can also manually tweak predictions based on homer picks and other things your leagues do. Lots of Niner fans in all my leagues so Niners get a bump in predicted pricing/ADP. But when you bump one group of players, you need to bump down another group to offset everything. Also some years, specific player tiers are more heavily loaded than others. For example, this year, starting QBs after the top 3-4 are all pretty much in one tier and that should be reflected in the predictions. For pricing I'll probably bump up the QB8-12 spots a bit higher than previous years. I'll probably offset this a little amongst the top 3 QBs, and also at other positions.
Now if your league does weird things like nominating kickers early or nominating a lot of late round players early, this can have an effect on the pricing and might make this more difficult. Luckily, mine doesn't.
Here is how my QB predictions went last year, You can see my estimate vs the actual player price. (Our league uses a lower cap than the standard $200.)
This works for snake drafts too. Just put in your draft pick history against that year's ADP and look at any patterns. I've learned the following about my snake leagues.
- In one of them, QBs get drafted a full round earlier than ADP on average and RBs about a half round later than ADP once you get past the top tier.
- In another, a ton of the other drafters pick their defense before they start drafting any backup players. I always knew defenses went too soon, now I know at what point in the draft the defense runs will start.
- One guy in my league went to college with Aaron Rodgers and knew him and always drafts him. Last year, the poor sap drafted him in the first round (superflex).
