by Jim Perrone a.k.a. "Foursport"
FORWARD by Greg Alan
Henry Bass, Brandie Searle and the entire team at pcDrafter have built one heck of a fantasy football software tool.
Below, Jim Perrone offers up a groundbreaking analysis showing how GOLD beats Average Draft Position (ADP) and in turn the highly correlated Value Based Drafting (VBD) approach on draft day. The powerful properties of the GOLD algorithm and Jim's real world analysis are very convincing.
Drafting a team that out scores even your best opponents by 10% or more on Draft Day is huge!
Jim, who typically competes in 20+ leagues a year, including WCOFF (Vegas) and the FFTOC (Online Championship), shows us how it works.
But enough of my writing. It's now time for you to read Jim's work, see the documented benefits and go get 'em.
Remember your first time? I do - it was late August 1993 and I had been invited to join a 'fantasy football league' by a co-worker. I had no idea what I was getting into but showed up on a Friday night armed with a six pack of O'Douls (no drafting drunk for this guy) and a newspaper (remember those?) and saw the most amazing pick of my fantasy football life right there in front of me and which has not yet been topped - 'and with the 2nd pick of the 2nd round - I select the Cleveland Brown Defense'. NO - that was not me!
Flash forward 13 years to the present season. No longer am I in 'TD only leagues' - I am playing in leagues where we go down to tenths and one hundredths of points in scoring systems so detailed that, if it were not for web-based league management systems, we'd still be adding up the results from last year. Oh - and playing in one league, you can write that off as ancient history as this year I have reached the 25 league mark. Before I continue, let me more formally introduce myself to you - 'my name is Jim and I am a Fantasy Football Addict'.
In order to play in 25 leagues, I would need something more than a newspaper and a fantasy football magazine. Years ago, I found software that had the players in a database and I was now able to make sure that I didn't embarrass myself by attempting to draft the best player in football in the 12 th round of a 16 th round draft. The software also helped me track the draft for later input into whatever website the league was using to run stats and track standings. For a few years, this was pretty much the only thing that this program could do.
What eventually replaced this simplistic draft tracking software were programs that incorporated the concept of loading prior year's stats so that owners could now look at how a player performed LAST YEAR as a guide to how they should draft THIS YEAR. Then came projecting this coming year's stats. That was an immense help. You could now see how your players might stack up against each other and start making decisions based on more current, and therefore useful, data. The growth of fantasy football and the widespread use of the Internet as a tool for pulling us fanatics together soon thereafter led to 'mock drafting' and the concept of the Average Draft Position (ADP).
To summarize the ADP drafting method in a nutshell - the concept is simple and states that if we gather lots and lots of data on what other fantasy football players are drafting in their leagues then we will see trends develop which will help us better figure out where we should draft any given player in our leagues. ADP is a very useful tool but, as you will see shortly, it has it drawbacks chief among them being the fact that 'mock drafts' only seem to be serious for the first 4 or 5 rounds and then many drafters drop out and let the computers pick for them. I am highly skeptical to say the least of data that is driven by unknown people of unproven knowledge or skill or, worse yet, computer programs. There is also a secondary problem inherent in the fact that many ADP lists are far too inclusive of all kinds of drafts and don't screen well enough to tease out the various types of re-draft leagues, keeper leagues and/or dynasty leagues that are contributing to the overall ADP numbers. If the ADP list has included 50 redraft leagues where a player is drafted in the 9th round - 12th pick (#108) but 75 dynasty leagues where the same player is drafted 4th round - 6th pick (#42) - his ADP slot will end up sliding into the 6th round - 8th pick (#68) - far too high to be useful in most redraft leagues.
Another popular method of attempting to draft a well oiled fantasy football machine is to look at Value Based Drafting (VBD) where it is clearly acknowledged that not all positions in fantasy football are equal, are weighted differently and therefore should be drafted accordingly. There are a limited number of high scoring running backs compared to say wide receivers so running backs are scarcer and more valued in most scoring systems and leagues. This recognition helped spur VBD to overcome ADP drafting for quite a while but the popularity of VBD has so changed the drafting environment that ADP now very closely matches VBD principles and, in many cases, is virtually indistinguishable.
SO - if everyone knows ADP and VBD principles, what can you do to set yourself apart from the competition?
THE ANSWER - pcDrafter GOLD!
This is direct from their website homepage:
'GOLD builds on Value Based Drafting principles. These ideas have been in the public domain for years. However, GOLD goes far beyond as it tightens up the process and dynamically delivers more robust player recommendations. This results in you drafting a vastly improved team.'
Well, I always say 'PROVE IT.'
I decided to take this little piece of software for a ride - you know, kick the tires so to speak and see if it could deliver.
In 'real life' I am a Certified Public Accountant and crunch numbers for a living - I also believe that winning fantasy football is all about the numbers. Too many fantasy owners fall in love with a stud name and the hype that surrounds a player (can you say Reggie Bush?). They lose sight of what winning is all about - simply outscoring your opponent no matter who is on your team! Like all fantasy football owners, I have favorite teams and favorite players when I sit down to watch the games BUT I have to put all of that aside come draft day or I will have a sub-par draft - so will you!
If you decide to stop reading the article here before it gets into all of the number crunching - that's fine with me as long as you remember this one point - pcDrafter GOLD delivers on its promise!
And it's not only what it does to improve your team but also the impact that it has on the other teams that will have you singing the praise of this program all season!
For the details of WHAT the program does and HOW it does it - please visit www.pcdrafter.com BUT if you want know how your team will turn out - read on!
My goal was to duplicate many of the common leagues out there in the fantasy football world. If I fell a little short of your league, don't worry as this program is easily adaptable to your league no matter its size, roster requirements and scoring rules. But for my testing system, I assumed the following:
- 12 teams
- 16 rounds
- Starting lineup of 1 Quarterback (QB), 2 Running Backs (RB), 3 Wide Receivers (WR), 1 Tight End (TE), 1 Kicker (K), 1 Defensive Team (DT) and 1 Flex position (RB, WR or TE).
- Performance based scoring - 4 points for passing touchdowns and 6 points for all receiving and rushing touchdowns. Passing yardage is at .05 points per yard and rushing and receiving yardage got players .1 points per yard. I also used the one point per reception rule since this seems to be growing in popularity - kicker and defensive scoring was pretty standard. I also assumed negative 2 points for interceptions thrown and fumbles lost.
- I limited all teams to drafting exactly 2 QB, 5 RB, 5 WR, 2 TE, 1 K and 1 DT as I wanted all teams to draft equally and see what happened.
With those parameters in mind, I set out on my task to compare drafting a league with just ADP info and then redrafting that entire league where all the other teams used ADP but I utilized the GOLD method to choose my team.
My team was cast in the 11th draft spot. WHY? Well, that's a tough spot to draft out of and I have that draft spot in one of the big contests this year. Since I was going to do all of this work, I figured I would double dip and get some personal learning out of it as well.
Now, to the drafts themselves.
ADP DRAFT RESULTS
This was a real challenge as I wanted yesterday's ADP figures and did not want keeper and dynasty leagues to contaminate and skew the listing by overvaluing certain players. I finally was able to track down an ADP list that was only 1 day old when I wrote this and had info on only redraft leagues that had drafted in the last 14 days. Another key factor was that these were from real drafts for serious money and not those mock drafts I questioned earlier.
I fired up pcDrafter GOLD and went to work entering ALL teams into the system using ADP and ADP only. I made no deviations from the list except when roster limitations dictated that I should - i.e. - the team already had its 5 WR limit.
When the draft was over, I studied all of the rosters looking at the projected points for each team in the league and used a points only approach to evaluating the league and my team. I examined both total team strength (based on the full 16 man roster) and the strength of a projected starting lineup (best available starters based on lineup criteria)
Drafting out of that 11th spot, I ended up with a total roster strength that would put me 5th in the league but, if you looked at starters only, I ended up with the 8th best team. I was strongest at the TE and WR positions and a little lower than average with my RB and QB positions. I did pretty well with my K but not too good at the DT spot. Here's a look at how the teams fared by draft position based on total points and then average points (based on a 16 week NFL schedule)
That's my team JP with highlighting - not too good!
For those of you picking out of the #1 or #2 spot - I can see why maybe you can find it in your heart (and your pocketbook) to stick with pure ADP drafting - but not me - I need a better system.
Whether it's $1 and pride or $1,600 or more on the line, I know that I want a better than 8th place finish out of my team.
Take the data above - study it position by position and draw your own conclusions on how ADP works or doesn't.
Next, I'll focus on the GOLD draft and discuss what happened that was different and why I will trust this product with all 25 of my leagues this year!
GOLD - THE NEW STANDARD IN HUMILIATING YOUR LEAGUE
Ok - I agree that may be a little over the top. Maybe you don't want to humiliate them since you will want them to come back so you can take their money again next year (and the year after that.) :-)
First, I cleared out the draft inside of pcDrafter and started over.
This time I continued to use the ADP method to draft for the other 11 teams BUT inputted the GOLD program recommendations when making picks for my team in the 11th spot.
While I was tempted to second guess the picks, this was a test and, as it turned out, I am extremely glad that I did not.
Let me jump right to the results and then discuss what I saw the GOLD program do differently from ADP.
So with no further ado, let me introduce you to TEAM JP - the best team in the league both in TOTAL STRENGTH and STARTER STRENGTH.
What an improvement for me and what a solid and convincing testimonial for pcDrafter GOLD!
The #1 draft spot is still a sweet spot to draft out of even using the ADP method but draft spots #2 and #5 did very poorly once I (and GOLD) messed with them by getting better value out of the GOLD picks. And, to overstate the obvious - stay out of the #12 spot if you use ADP - bad news there.
If you have the #12 spot in your league - grab GOLD or just kiss your money goodbye. As I said earlier, it's not only what this program will do for YOUR team, it's what it forces the OTHER teams to do as well.
As you can see, I ended up being the top scoring team with 140.38 average points per week which is 6 more points per week then the closest competitor. Yes - there will be bye weeks and such to take into consideration (which I did not do here) but this is a team that I could easily project as being 9-5 or even 10-4 over a 14 week fantasy football season and that should be top notch enough to carry me into the playoffs.
Here's a chart that highlight the changes that happened due to drafting with GOLD as your advisor rather than relying on ADP.
In this chart, you can see using GOLD during the draft changed each and every team - sometimes for worse and sometimes for better. Wherever you see 1.00, this means that nothing changed at all for the team. They ended up in exactly the same situation they were in before - usually meaning that they had the same players at many positions.
Where the numbers are less than 1.00, this is an indication of the teams percentage of their former strength - so in the case of team #2, they are only 89% of the team they were before GOLD started helping me draft better and changed how they had to draft as a result.
In those areas where the numbers are greater than 1.00 (see my TEAM JP) you are looking at improvement that pushes my team to110% of the team it was under the ADP method. SWEET!
7 of the 12 teams in this league drafted worse (58%) while 5 got better BUT no team had a larger increase than my team using GOLD. I got significantly better at WR and TE and that is a huge difference maker in a league with a flex position. I wish I had done better a RB but you can't have everything and drafting out of the 11th spot typically means you really only get a shot at one good RB anyway.
GOLD definitely finds value for teams that invest in using the pcDrafter program. It guides you towards performance (remember that it's the numbers that win) and away from hype.
Occasionally GOLD would recommend the same player as the ADP method but, in many cases (approximately 45-50% of the time), I noticed that players being drafted with the ADP method were being taken anywhere from 8-10 picks earlier than GOLD recommended. When it got to those crucial season making or season breaking mid-rounds (6-10), the ADP picks were coming 19 or 20 picks too early based on the GOLD system.
This is great news - if your leaguemates are drafting with ADP in mind - they will be overextending themselves by taking guys too soon and will leave even more value for you to scoop up on both now and later.
VERY EASY TO USE
The ease of setup and use is astonishing - I haven't covered this here in this article but I thought I should throw that out there. You can find tips and tricks in either the help file or online. I have used other software for years and I found that I adapted easily to pcDrafter. Have no fear - you will be up and running (and drafting to win) in pretty short order.
To summarize (in case you jumped down here instead of reading the entire article and studying the charts) - pcDrafter GOLD vastly improved my team when used instead of typical ADP approaches. Further, GOLD very adeptly countered ADP drafting. GOLD has a killer punch and is very effective at counter punching as well.
I went from being an 8th place loser to a 1st place contender.
GOLD drafting solidly and consistently showed me where value could be had and kept me from overreaching for less productive players. The ability to tweak all of the different aspects of your league will prove invaluable to all drafters. Using pcDrafter and the GOLD algorithm will help you draft a stronger, balanced and more competitive team
Happy Drafting - SEE YOU IN THE PLAYOFFS!