Using Probability to Find Fantasy Bargains
Professional Poker Player Liv Boeree did a TedX presentation discussing probability and how it influences decision-making in our daily lives. For example, if you tell your wife you were likely to be home from the Pub with your friends at 8:00 PM, it is an almost certainty that she is expecting you home by 8:00 PM. My wife purposely sets all of the clocks forward in my car and on my phone because, to me arriving at X:XX means arriving at X:XX + 15 minutes.
Boeree’s presentation got me thinking about blanket statements, assumptions and perceptions that apply to how we think about fantasy football. Most often, we make assumptions that greatly discounted players occur for unlikely reasons. In short, instead of discarding important pieces to our championship teams, we move towards a probable outcome that deflates their value.
Some common reasons for this Discount/Discard:
- Injury (Rob Gronkowski in 2015, Jimmy Graham, Hakeem Nicks, Crabtree, Cruz, Julio Jones)
- Age (Frank Gore, Demarco Murray, Calvin Johnson, Antonio Gates, Larry Fitzgerald)
- Perceived fits or assumed targets on offense (Eric Ebron and Jack Doyle, O.J. Howard and Cameron Brate, Trey Burton and Adam Shaheen – notice a trend with the TE’s here?)
- Recency Bias – (Todd Gurley pre-2017)
- Perception bias (personal bias, or origin bias – Will Fuller, Chris Godwin and John Ross)
When we think of game theory and hit rates, we are looking for a safe outcome. Dynasty research tells us first round rookie draft picks are historically more likely to hit than second round draft picks. According to the chart above, the probability of a second round pick beating out our first round rookie draft picks is at least 40-60% of the time. We are not perfect, as drafters, or as owners.
All that has to happen for us to be successful – whether drafting, trading, or adding players to our waiver wire, is to have players beat our expectations. If we feel like Adam Shaheen’s role is not going to be a significant asset in Chicago’s offense this year, we have a 15-40% chance of being right. If we pay a 15-40% price for Shaheen and he surprises us with an even outcome, we have automatically profited at least 20% from our purchase price. So what’s the recommendation? If owners are scared of Shaheen’s new situation and you can buy from a scared owner – you have a high probability (60-85%!) to significantly benefit.
So how can I start applying these “hit rates” today?
Using Jacksonville’s wide receivers, we can look to see how owner discounts and bias/perception have led to potential price bargains.
Jacksonville’s 2018 WR core –
Shared Red Flag Discounts:
(These are reasons that owners/drafters through either research, history or preference, discounted the value of all Jaguar WRs.)
- Blake Bortles
- Offensive Scheme with Run-heavy emphasis
- Strong Defense (game flow)
- Unsustainable Garbage Time Stats that inflated stats (2015-2016)
- Crowded, Unclear Depth Chart
- Blake Bortles (seriously)
Keelan Cole – ADP 147, WR ADP 59 (WR 5)
|2017||WR 45 (91 fantasy pts)||16/6||83||42||748||3||46.8||50.6|
Red Flags/Discount Thoughts
- Competition (not viewed as primary receiver)
- Small college pedigree (Kentucky Wesleyan)
- Undrafted free agent, low draft profile
- Possibly THE reason Jaguars released Allen Hurns
- Between weeks 13-15, Cole put together 393 yards (186 yards in week 14!, Averaging 131 Y/G)
- 6/16 games played Cole had seven or more targets
- Hyper efficient and under the radar production
Dede Westbrook – ADP 118, WR ADP 55 (WR 4.5/5)
|2017||WR 98 (40 fantasy points)||7/5||51||28||339||1||48.4||52.9|
- Injured reserve (September 8 – November 18)
- Late draft pick investment
- Rookie age (23-24)
- Lack of consistent production (Only three games over 42 yards)
- Westbrook did not start until Game 10, but averaged six or more targets 5/6 games played
- Has Bortles trust as a go-to option
- Looked explosive with 12.6 Yards per Reception
Donte Moncrief – ADP 134, WR 46 (WR 4/5)
|2017||WR 80 (51 fantasy pts)||12/8||47||26||391||2||32.6||55.3|
|2016||WR 83 (73 fantasy points)||9/7||56||30||307||7||34.1||53.6|
|2015||WR 38 (109 fantasy points)||16/10||105||64||733||6||45.8||61.0|
|2014||WR 76 (64 fantasy pts)||16/3||49||32||444||3||27.8||65.3|
- Lack of consistent production in Indianapolis
- New offense, with target competition
- Injury history (21/32 games played in last 2 years)
- 2017 and possibly 2016 should be considered a mulligan with Andrew Luck out
- One year “Prove It” contract with the Jaguars
- 3rd round draft pedigree (90th, 2014 draft)
- High red zone efficiency
- May have the most clear role as a deep speed target (4.34-40 yard time)
Marqise Lee – ADP 137, WR 57 (WR 5/6)
|2017||WR Rank 43 (Fantasy Points 92)||14/14||96||56||702||3||50.1||58.3|
|2016||WR 44 (109 fantasy pts)||16/6||105||64||851||3||53.2||60.0|
|2015||WR 114/ (29 points)||10/1||32||15||191||1||19.1||46.9|
|2014||WR 96 (49 fantasy points)||13/8||69||37||422||1||32.5||53.6|
Red Flags/Discount Thoughts
- Widespread rumors that he would be cut in the offseason and not Hurns
- History Bias: He came into the league with the aforementioned players, but heavily overshadowed by Allen Robinson
- High draft pedigree (2nd round, 39th pick), Allen Robinson drafted second round, 61st pick
- Durable: has played 29/32 games in 2016-2017
- Target favorite: With 211 targets, he is averaging 7.3 targets per game
- May have most clear path to WR 1 role in Jacksonville indicating high reward/return
(Bonus) Exiting Jaguar Players:
Allen Hurns (FA) – ADP 229, WR 83 (WR 7!)
|2017||WR 67 (62 fantasy pts)||10/8||56||39||484||2||48.4||69.6|
|2016||WR 77 (66 fantasy points)||11/11||76||35||477||3||43.4||46.1|
|2015||WR 14. Overall 35. (163 fantasy points)||15/16||105||65||1031||10||68.7||61.0|
|2014||WR 44 (104 fantasy pts)||16/8||97||51||677||6||42.3||52.3|
Allen Robinson (CHI) – ADP 25, WR 15.75 (WR 1/2)
|2017||N/A (2 pts)||1/1||1||1||17||0||17||100|
|2016||WR29 (126 pts)||16/16||151||73||883||6||55.2||48.3|
|2015||WR 4/ 7th Overall (224 points)||16/16||151||80||1400||14||87.5||53.0|
|2014||WR 70 (69 pts)||10/8||81||48||548||2||54.8||59.3|
Conclusions from a fantasy outlook…
Looking at current Jaguar WRs, we have an interesting mix with very little price/cost separation between Marquise Lee, Donte Moncrief, Keelan Cole and Dede Westbrook. Using the ADP chart from above, Drafters are only “certain” on Allen Hurns and Allen Robinson. So, what can we make of this? Using Boeree’s Probability Outcomes, when we trade, draft, or acquire one of these players we ONLY have to make sure that the cost is cheaper than the expected value return. So what are the costs?
Cost: Mid second round start up value, with Expected Return of a WR 1 outcome
Probability of reaching this cost: About Even, or 40-60%. There is some owner preference and subjectivity here, but this is what you need to decide using Boeree’s Probability Chart.
Action: Sell. If there is a 40-60% chance he returns to his 2015 season (my preference), I believe he will finish below his WR 1 price and owners would be safer selling than banking on production at current price.
Cost: 18th round start up value (ADP 229), with expected Return of a WR 6-7 outcome
Probability of reaching this cost: Most Often to Very Likely (80%-95%).
Action: Buy. His price is extremely cheap and, given what I have seen, I expect Hurns to rebound into WR3 territory which is a huge return on investment.
Donte Moncrief, Marquise Lee, DeDe WestBrook, Keelan Cole
Cost: 12th-13th round start with little separation between all four players
Probability of reaching this cost:
Donte Moncrief – Likely/Probably (about 60-85%)
Marqise Lee – Most Often (80-95%).
Dede Westbrook – About Even (40-60%)
Keelan Cole – About Even (40-60%)
Strong Buy – Marqise Lee.
Buy: Donte Moncrief.
Hold or Sell: Westbrook and Cole.