Predict auction draft prices by simulating the entire information ecosystem.
The idea
We built a simulated world. Inside it, 20 fantasy drafters and 16 sports media personalities actually exist — each with a full life history, career, personality, biases, information diet, and memories from past drafts. The analysts post takes. The influencers post hype. The beat reporters post practice notes. The rumor accounts post whispers. And the drafters? They scroll through it all, each one seeing different content based on who they are and what they follow.
Then they sit down at a 16-person auction table and bid on players using the beliefs they formed from everything they consumed. The prices that emerge aren't assigned — they're emergent.
That's DraftFish.
It's built on MiroFish, a multi-agent social simulation engine that predicts real-world events by simulating how information flows through society. MiroFish's thesis: simulate the information ecosystem accurately enough, and predictions emerge from agent behavior. You don't hard-code what people think — you simulate what they see, and their beliefs follow.
What DraftFish adds
MiroFish simulates Twitter and Reddit to predict events. DraftFish uses the same engine but adds three layers:
Sports media ecosystem — analysts, influencers, beat reporters, and rumor mills that produce content grounded in real player data
Drafter personas with life histories — not "you are a conservative drafter" but a full backstory that produces conservative behavior through accumulated experiences (inspired by Stanford's Generative Agents)
Auction engine — beliefs from the social sim feed into a rules-based auction that runs thousands of times to produce price distributions
ETP = Points Per Game × Expected Games
A player scoring 25 PPG on a 6-seed (expected ~2.1 games) has lower ETP than one scoring 15 PPG on a 1-seed (expected ~4.5 games). The tournament structure matters more than raw scoring.
How it works
🏆
Real Data
108 players, odds, stats
→
👤
Personas
20 drafters, 16 media
→
💬
Social Sim
96 posts, 321 reads
→
🧠
Beliefs
Memories → valuations
→
💰
Auction
400 simulated drafts
→
🎯
Value
ETP vs market price
The social sim is the core. It's not overhead — it's the belief formation mechanism. Without it, you'd have to assign what each agent thinks. With it, beliefs emerge from what each agent was exposed to.
Reading the numbers
Every player gets four numbers:
ETP (Expected Tournament Points) — the total points we predict this player will score across all tournament games. Accounts for PPG and how many games their team is likely to play.
Mkt$ — what the simulated room paid for this player (median across 400 auctions). This is what you should expect to pay in a typical room.
Value — ETP minus market price. Positive = the room underprices this player. You're getting free points. The higher the number, the better the deal.
Range (P10-P90) — what they cost in a cheap room vs an expensive room. Wide range = volatile pricing, opportunities to steal.
Using this tonight
Before the draft: Know your targets. The Value column is your cheat sheet. Focus on players with high ETP and low Mkt$ — those are the ones casual rooms won't fight over.
During bidding: The Mkt$ is your anchor. If the room bids a player past their Mkt$, let them go — someone else is overpaying. If they're going below Mkt$, bid confidently.
After each pick: Diversify teams. If you already have a Duke player, be cautious on another Duke player — one upset eliminates both. Spread across 4+ teams.
Pick 6 (bench): Spend $1-4. Your 6th player only matters when a starter is eliminated. Grab a cheap player on a 1-2 seed who'll have games left to score.
Budget pacing: Spend 70-80% of your $100 on your first 4 picks. Don't panic-save and don't panic-spend. The sim shows most value is in picks 2-4, not pick 1.
What the simulation found
Across 400 simulated auctions with 20 behaviorally diverse personas, the market told a clear story:
Depth players on 1-seeds are systematically underpriced. Michigan's 3rd-4th scorers cost $4-5 with ETP comparable to Dybantsa at $26 — a 5x premium driven entirely by name recognition and influencer hype.
Why? Because casual agents consumed influencer content that amplified star names, while depth players were invisible to them. The analytics-oriented agents saw the KenPom data and valued depth correctly — but they're outnumbered in a typical room. The information asymmetry IS the market inefficiency.
Suggested roster ($100 budget)
Optimized from sim data — maximizes total ETP while diversifying across teams and keeping budget for bidding wars.
These players have 40+ ETP but the simulated room barely notices them. If you can get any of these for $1-5, you're stealing points.
Aday Mara(Michigan) 51 ETP for $5Tobe Awaka(Arizona) 46 ETP for $4Rueben Chinyelu(Florida) 46 ETP for $4Pryce Sandfort(Nebraska) 46 ETP for $4Walter Clayton Jr.(Florida) 45 ETP for $4Elliot Cadeau(Michigan) 45 ETP for $4Tamin Lipsey(Iowa State) 43 ETP for $4Nick Boyd(Wisconsin) 44 ETP for $5
Overpay traps — don't get caught
These players get bid up well past their ETP because they have name recognition, hype, or both. Let someone else overpay.
Thomas Haugh(Florida) $30 but only 70 ETPBrayden Burries(Arizona) $26 but only 70 ETPDarius Acuff Jr.(Arkansas) $26 but only 59 ETPAJ Dybantsa(BYU) $26 but only 52 ETPIsaiah Evans(Duke) $25 but only 66 ETPGraham Ike(Gonzaga) $22 but only 55 ETP
Full rankings
Value = ETP - median simulated auction price. Positive = the market underprices this player.
#
Player
Team
Sd
ETP
Mkt$
Value
Range
1
Cameron Boozer
Duke
1
100.4
$36
+64.4
$32-$41
2
Morez Johnson Jr.
Michigan
1
57.6
$10
+47.6
$5-$20
3
Koa Peat
Arizona
1
63.6
$17
+46.6
$10-$25
4
Jaden Bradley
Arizona
1
59.2
$13
+46.2
$7-$22
5
Aday Mara
Michigan
1
51.0
$5
+46.0
$2-$8
6
Emanuel Sharp
Houston
2
52.3
$7
+45.3
$4-$12
7
Alex Condon
Florida
1
51.2
$6
+45.2
$3-$11
8
Yaxel Lendeborg
Michigan
1
64.2
$20
+44.2
$11-$29
9
Keaton Wagler
Illinois
3
50.8
$7
+43.8
$3-$12
10
Brayden Burries
Arizona
1
69.7
$26
+43.7
$16-$32
11
Milan Momcilovic
Iowa State
2
55.7
$13
+42.7
$7-$22
12
Kingston Flemings
Houston
2
55.7
$13
+42.7
$7-$22
13
Tobe Awaka
Arizona
1
46.0
$4
+42.0
$2-$8
14
Rueben Chinyelu
Florida
1
45.9
$4
+41.9
$2-$7
15
Pryce Sandfort
Nebraska
4
45.7
$4
+41.7
$2-$8
16
Walter Clayton Jr.
Florida
1
45.1
$4
+41.1
$2-$7
17
Isaiah Evans
Duke
1
65.9
$25
+40.9
$15-$34
18
Elliot Cadeau
Michigan
1
44.9
$4
+40.9
$2-$7
19
Thomas Haugh
Florida
1
70.0
$30
+40.0
$19-$35
20
Tamin Lipsey
Iowa State
2
43.3
$4
+39.3
$2-$7
21
Nick Boyd
Wisconsin
5
44.1
$5
+39.1
$2-$10
22
Jeremy Fears Jr.
Michigan St.
3
43.9
$6
+37.9
$3-$12
23
Labaron Philon
Alabama
4
55.7
$18
+37.7
$10-$26
24
Braden Smith
Purdue
2
44.0
$7
+37.0
$4-$15
25
Kon Knueppel
Duke
1
53.1
$18
+35.1
$8-$30
26
Caleb Foster
Duke
1
44.2
$10
+34.2
$4-$18
27
Graham Ike
Gonzaga
3
55.0
$22
+33.0
$11-$33
28
Darryn Peterson
Kansas
4
50.8
$18
+32.8
$9-$27
29
Darius Acuff Jr.
Arkansas
4
58.7
$26
+32.7
$17-$32
30
AJ Dybantsa
BYU
6
51.7
$26
+25.7
$17-$35
31
Zakai Zeigler
Tennessee
6
25.5
$9
+16.5
$6-$13
32
Donovan Dent
UCLA
7
26.4
$10
+16.4
$7-$15
33
Larry Johnson
McNeese
12
26.4
$10
+16.4
$7-$15
34
Christian Hammond
Santa Clara
10
25.3
$9
+16.3
$6-$14
35
Pete Suder
Miami (OH)
11
24.2
$8
+16.2
$5-$12
36
Tyler Perkins
Villanova
8
23.2
$7
+16.2
$4-$11
37
Cruz Davis
Hofstra
13
25.1
$9
+16.1
$6-$13
38
Terry Anderson
High Point
12
24.1
$8
+16.1
$5-$12
39
David Punch
TCU
9
23.1
$7
+16.1
$5-$11
40
Izaiyah Nelson
South Florida
11
26.0
$10
+16.0
$6-$15
The information ecosystem
Below is what actually happened inside the simulation. 16 ecosystem agents posted content across 6 rounds — analysts, influencers, beat reporters, rumor accounts. Each has a distinct voice, backstory, and bias. Drafters selectively consumed posts based on their information diet — an ESPN-only casual never sees the KenPom thread, and the analytics nerd never sees the Instagram hype reel.
T
TheFilmRoom_Donovan@000·analyst
Just spent 3 hours rewatching Cameron Boozer film and I need to talk about something nobody's mentioning. The 22.7 PPG is the headline but watch his footwork in the mid-post. Textbook drop step. He sells the pump fake like a veteran — draws contact, finishes through it. This is a kid who was coached RIGHT. Duke's 1-seed is built on fundamentals most pros never develop. Thread coming tonight. #FilmRoom #NCAATournament
#Cameron Boozer#Duke
💬 14🔁 2❤️ 166👁 84K
K
KenPom Kommand@001·analyst
Hot take grounded in data: AJ Dybantsa's 25.3 PPG at BYU (6-seed) is impressive but schedule-adjusted efficiency tells a more complicated story. Darryn Peterson's 19.8 for Kansas against B12 defenses night in and night out is underrated by comparison. Both are first-round locks. The question is which one sustains production against elite tournament defense. The KenPom splits don't lie. Full newsletter breakdown dropping at 8PM. #KenPom #NBADraft
#AJ Dybantsa#BYU#Darryn Peterson#Kansas
💬 31🔁 8❤️ 195👁 47K
S
SeedAdvantageAnalytics@002·analyst
Predictive model output for this NCAA field: Four 1-seeds (Duke, Florida, Arizona, Michigan) hold 9 of the projected top-15 picks. That concentration is historically anomalous. Cameron Boozer, Thomas Haugh, and Yaxel Lendeborg all projecting top-5. My selection committee experience says this bracket is top-heavy in a way that creates 5-12 chaos. 1-seeds advancing to Final Four: 67% in current model. Posting full probability tables now. #BracketBreakdown #Analytics
#Cameron Boozer#Duke#Thomas Haugh#Florida
💬 4🔁 16❤️ 265👁 187K
R
ReadTheRoom_Sports@003·analyst
Everyone running their PPG numbers and I'm over here watching Jaden Bradley make guards miserable for 35 straight possessions. 13.5 points per game. That number means NOTHING. I coached defense for 15 years. I know what I'm looking at. Bradley is the reason Arizona is a 1-seed — not despite his scoring line, but because of what he does without the ball. Kids today want to count buckets. I count stops. Arizona is dangerous. #DefenseWinsChampionships
#Jaden Bradley#Arizona
💬 25🔁 9❤️ 38👁 87K
T
TheFilmRoom_Donovan@000·analyst
Spent the morning re-watching Arizona's offensive sets from their last three games. Jaden Bradley's court vision is genuinely special — he's reading second and third options before the ball is even inbounded. But what jumps off the screen is how Koa Peat plays off his gravity. When Bradley collapses the defense, Peat's footwork in the dunker spot is as clean as any big I've seen at this level. The Wildcats are a dangerous 1-seed because they move the ball with purpose, not just talent. Round 2 opponents need to account for all three of Burries (15.9 PPG), Peat (14.5), and Bradley (13.5) — you can't scheme for one without opening a window for the other two.
#Jaden Bradley#Koa Peat#Brayden Burries#Arizona
💬 18🔁 2❤️ 59👁 84K
K
KenPom Kommand@001·analyst
Running the numbers ahead of Round 2: Cameron Boozer at 22.7 PPG for a 1-seed Duke squad is a strong statistical profile, but context matters — AJ Dybantsa at 25.3 for a 6-seed BYU tells a very different efficiency story. Leading scorer on a 6-seed hitting those numbers means he's doing it with weaker surrounding talent and against a compressed schedule. The adjusted efficiency gap vs. tournament-caliber competition is the real question. Meanwhile Michigan's load distribution — Lendeborg (14.6), Johnson (13.1), Mara (11.6) — gives the Wolverines the lowest scoring variance of any 1-seed in the bracket. That's a KenPom-friendly profile.
#Cameron Boozer#AJ Dybantsa#Yaxel Lendeborg#Duke
💬 30🔁 16❤️ 8👁 47K
S
SeedAdvantageAnalytics@002·analyst
Historical seed performance models flag two patterns for Round 2: 1-seeds with a 20+ PPG primary scorer survive at a 73% clip — Boozer (22.7, Duke) fits cleanly. But the model also consistently over-indexes 4-seeds with elite guard production. Darius Acuff Jr. (22.9 PPG, Arkansas) and Labaron Philon (21.7 PPG, Alabama) both match a historically dangerous archetype. Arkansas projects at 36% upset probability against a 1-seed matchup. The outlier play: Graham Ike (19.7 PPG) at 3-seed Gonzaga has the second-best efficiency-adjusted scoring profile in the entire field. My model likes him more than his seed suggests.
#Cameron Boozer#Darius Acuff Jr.#Labaron Philon#Graham Ike
💬 19🔁 17❤️ 23👁 187K
R
ReadTheRoom_Sports@003·analyst
I'm not buying the Duke hype on Boozer's 22.7 alone. What I want to see is how Isaiah Evans holds his assignment on the weak side when the ball reverses — that's where the defensive breakdowns will show up. Kon Knueppel (12.0 PPG) is a shooter, but can he fight through a hard screen three possessions in a row? Duke is gonna face physicality in Round 2 and I need to see their wings hold ground. Flip side — Florida's Thomas Haugh (17.1) plays hard every single possession. That relentless effort doesn't always pop in the stat line but any coach watching that film sees it immediately. That's a winning player.
Watched 6 hours of Cameron Boozer tape last night and I keep coming back to the same thing — his footwork in the post is unlike anything I've seen from a freshman in years. As a former receiver coach, I know what it looks like when someone has been drilled on their footwork since age 8. Boozer has that. The drop step, the pivot, the finish through contact — all there at 22.7 PPG. Duke's 1-seed isn't about the bracket. It's about him. Film don't lie. #MarchMadness #TheFilmRoom
#Cameron Boozer#Duke
💬 7🔁 9❤️ 298👁 84K
K
KenPom Kommand@001·analyst
Running the numbers on Round 3 matchups. AJ Dybantsa (25.3 PPG) as a 6-seed is the single most dangerous offensive player in the bracket by usage-adjusted efficiency. BYU's schedule-adjusted numbers put them inside the top 20 despite the seed. Meanwhile Duke's supporting cast — Evans at 14.9, Knueppel at 12.0 — creates enough spacing to make Boozer's 22.7 PPG repeatable in tournament conditions. The data says don't sleep on a BYU-Duke collision. Full model drops Thursday. #KenPom #CollegeBasketball
#AJ Dybantsa#BYU#Cameron Boozer#Duke
💬 30🔁 3❤️ 144👁 47K
S
SeedAdvantageAnalytics@002·analyst
Historical check: 6-seeds beat 3-seeds in the Sweet 16 roughly 38% of the time when they carry a top-5 scorer in the field. AJ Dybantsa at 25.3 PPG qualifies. My models also flag Darius Acuff Jr. (22.9 PPG, 4-seed Arkansas) as structurally undervalued — his points-per-possession in high-leverage moments ranks 2nd among all tournament guards. The 1-seeds are formidable but this bracket has multiple seeding-based vulnerabilities. #BracketScience #SeedAdvantage
#AJ Dybantsa#BYU#Darius Acuff Jr.#Arkansas
💬 5🔁 15❤️ 145👁 187K
R
ReadTheRoom_Sports@003·analyst
Stop me if you've heard this before — everyone's watching the scorer, nobody's watching what happens BEFORE the catch. I've been studying how defenses set against Boozer and the tell is always in the weak-side help rotation. When it comes late, he scores. When it comes early, he kicks. Isaiah Evans at 14.9 PPG is the beneficiary of that late rotation more than anyone on that roster. From a defensive coordinator's lens — you stop Duke by stopping Evans first. Boozer is secondary. Don't @ me. #ReadTheRoom #MarchMadness
#Cameron Boozer#Duke#Isaiah Evans
💬 29🔁 18❤️ 12👁 87K
T
TheFilmRoom_Donovan@000·analyst
Film room observation that nobody's talking about: Isaiah Evans' off-ball movement is what makes Duke genuinely dangerous in March. I watched six straight possessions last night focused exclusively on his baseline cuts, and the defensive attention he commands is real. Cameron Boozer at 22.7 PPG gets all the headlines — rightfully so, that post footwork is legitimate — but it's the way Evans (14.9 PPG) forces rotations that opens everything up. Coached receivers for 11 years. I know what it looks like when a guy understands spacing at a structural level. That Duke system is dialed in. Full breakdown dropping tomorrow morning. #FilmRoom #Duke #MarchMadness
#Isaiah Evans#Cameron Boozer#Duke
💬 19🔁 8❤️ 290👁 84K
K
KenPom Kommand@001·analyst
Updated model inputs for Round 4. The 1-seed concentration this bracket is historically atypical — Duke, Florida, Arizona, Michigan all alive. What the raw numbers tell you: AJ Dybantsa at 25.3 PPG on a 6-seed BYU is the single highest-variance data point remaining. His scoring volume on a mid-major schedule creates legit projection uncertainty. Darius Acuff Jr. (22.9 PPG, Arkansas, 4-seed) is the other outlier — 4-seed scoring profile that belongs in the top line. Cameron Boozer (22.7 PPG) on the 1-seed is historically the safest correlation for a deep run. Numbers don't lie. They just don't always tell you when. #KenPom #Bracketology #TournamentAnalytics
#AJ Dybantsa#Darius Acuff Jr.#Cameron Boozer#BYU
💬 39🔁 9❤️ 272👁 47K
S
SeedAdvantageAnalytics@002·analyst
Predictive model refresh for Round 4: 1-seeds historically convert at elevated rates at this stage, but this year's field has abnormal variance at the 4-line. Darius Acuff Jr. (22.9 PPG, Arkansas) and Darryn Peterson (19.8 PPG, Kansas) are both 4-seeds posting scoring averages that belong on 1-seed rosters. My upset probability model flags both Arkansas and Kansas as statistically overperforming their seed. On the 1-seed side, Arizona's three-pronged attack — Burries (15.9), Peat (14.5), Bradley (13.5) — is the most balanced scoring distribution among top seeds remaining. Depth profiles like that trend positively in late-round simulations. Proceed accordingly. #SeedAdvantage #Bracketology #MarchMadness
Everyone's talking about who's going to SCORE. I'm talking about who's going to GUARD Cameron Boozer. I spent 15 years coaching linebackers and defensive schemes — I know what an unsolvable matchup problem looks like, and Boozer at 22.7 a game is exactly that. You go big, he's too quick. You go small, he posts. You double, Duke has Isaiah Evans and Kon Knueppel ready to shoot. And it's the same problem from Arizona's frontcourt — Koa Peat at 14.5 PPG is a physical mismatch that punishes you for being undersized. Defense wins tournaments. Somebody has to solve these problems or it's a short weekend. The tape is telling. #ReadTheRoom #MarchMadness
20 drafter personas, each with a life story that produces their draft behavior. Not "aggressive risk-taker" as a label — a specific person with specific experiences that make them aggressive in some situations and cautious in others.
16 were drawn for each simulation. Their beliefs formed from what they consumed in the social sim.
M
Marcus Webb
29 · High School Physical Education Teacher · 2 yrs fantasy
Knowledgemoderate
Strategybalanced
SourcesESPN, NBA App, The Ringer NBA Podcast, Basketball Reference basic stats, Twitter/X player follows
Risk
Posts seen20
Background
Marcus grew up in Columbus, Ohio watching NBA games every night with his father, a former semi-pro who drilled into him that you read a player by watching how he moves without the ball — not by staring at a stat line. He played point guard through a small Division III school in Ohio, studying kinesiology and learning to evaluate athleticism and effort as physical signals. A knee injury in his junior year ended any playing ambitions, but sharpened his eye for body language and conditioning. He distrusts numbers he cannot connect to something he has personally watched. His instincts feel earned, not inherited.
Philosophy
"I trust what I see. If a guy looks right — the way he attacks closeouts, how he moves without the ball, whether he's playing with energy — that tells me more than any percentage. I try not to reach, but I won't let a player I've watched dominate slip past me just because his numbers don't pop."
Going into the draft
I've watched Cameron Boozer three times this week. The way he reads post defenders before the catch — that's not something you see in a stat line, that's feel. He's the real deal. But I've been burned before loading up on one guy, so I'm setting a hard ceiling and I'm walking away if the room goes crazy.
Dybantsa's 25.3 a night is genuinely absurd and I'd be lying if I said I'm not interested, but every single person at this table has heard the same hype, which means the bidding is going to get emotional fast. And he's a 6-seed. One tough draw and that scoring disappears after two games. I'm in at the right price, not at whatever the room decides he's worth.
Two separate whispers about Arizona players not being right — Burries moving differently and some vague maintenance day situation involving a high-usage 1-seed scorer. I got burned once ignoring stuff like this. I'm not touching Arizona at market price tonight, and I'm cautious on any 1-seed big scorer until something clarifies.
Grew up in Akron, Ohio, third of four kids in a union household where sports were the common language and ESPN was always on after dinner. Never much for school beyond what got him to his journeyman's card. Learned to make decisions fast on job sites — no time to overthink, trust what you see, live with it. Lost a side business in his late twenties betting on a gut feeling and it stung, but it didn't make him cautious, it made him want to swing bigger to get it back. He respects confidence over calculation.
Philosophy
"You want the best players. Everyone knows who the best players are. Get two of them and don't stress the rest — someone's always available on the wire. Overthinking it is just an excuse for not having the guts to spend."
Going into the draft
Boozer is the guy. I've been hearing his name every single day this week, ESPN's got him consensus number one, 22.7 points a game on a 1-seed. You want to win this thing, you start there and you don't apologize for the price.
Stars carry rosters. I had Giannis and KD in year eight and I coasted. The scrubs behind them figured themselves out. Two scoring machines at the top and the rest doesn't matter as much as people think. Same logic applies here.
Dybantsa is dropping 25 a game — that's a number you can't talk yourself out of. Yeah he's a 6-seed, but scorers score. If I can pair him with Boozer I've got a team.
SourcesBBall-Index, Cleaning the Glass, Second Spectrum, Synergy Sports, The Athletic, internal franchise scouting reports, game film (self-reviewed)
Risk
Posts seen23
Background
Grew up in Durham, NC with a father who coached high school basketball and a mother who was a biostatistics professor at Duke. Absorbed both worlds simultaneously—his dad's eye for off-ball movement and his mom's demand for rigor. Played D3 basketball at UNC Asheville, solid starter, never athletic enough to go higher. Watched film obsessively as a player. Got an M.S. in statistics at NC State. Now consults for a mid-market NBA franchise on player evaluation models. Professionally trained to distrust gut instincts without data support, but four years of playing the game taught him that certain things—effort, positioning, motor—only appear on tape.
Philosophy
"I want eight guys I believe in over one guy I'm praying for. The metrics tell me who to look at; the film tells me if I trust what I'm seeing. I've watched enough tape to know box scores miss half the story, and I've seen enough variance in my own models to know concentration risk is real."
Going into the draft
ETP is the signal, PPG is the story people tell about the signal. Dybantsa at 25.3 a night sounds electric until you remember he's a 6-seed and my model has him at 51.7 expected tournament points — less than half of Boozer's 100.4. The market is pricing narrative. I'm pricing games played times points per game times survival probability. Those are not the same number.
I've watched Cameron Boozer on tape three times this week and I grew up twenty minutes from Cameron Indoor. The way he reads post defenders before the entry pass arrives is a tell for a guy who won't get flustered when the bracket tightens. Duke bias is real — I know it — but in this case the film is actually confirming what my model suggests. That's the only time I let the bias run.
Concentration risk is the tournament killer. One first-round upset and I'm watching two or three roster slots go dark at the same time. I need players spread across at least four different teams. The 'but they're a 1-seed' justification is the same reasoning that got that guy in my mock draft to go $58 on one player. I taped his roster to my monitor for a reason.
SourcesESPN, The Ringer NBA Podcast, NBA Twitter, Yahoo Fantasy News, Bleacher Report
Risk
Posts seen20
Background
Grew up in Memphis with a dad who coached high school basketball, framing every game as a story of grit, redemption, or heartbreak. Studied graphic design at University of Memphis, training himself to see the world through narrative and visual pattern. Early in his career he invested savings in a friend's food truck because the concept had a compelling story — it failed. The lesson should have been to check the numbers, but instead he decided he just needed to pick better stories. He's intuitive, emotionally intelligent, and drawn to narrative coherence over analytical rigor.
Philosophy
"Give me the guy with a chip on his shoulder — coming back from injury, just got traded, has something to prove. I'd rather blow half my budget on one elite player who controls the ball than spread money across five forgettable guys. Stars carry leagues. Scrubs are just noise control."
Going into the draft
Boozer is the story of this tournament. I've watched him three times this week and there's nothing in that film that looks like a guy who loses in March. The way he reads post defenders before the entry pass — that's not a freshman, that's a closer. Duke's running a four-deep scoring rotation around him and they're a 1-seed. I'm spending big, same energy I had on Giannis.
Stars carry leagues, scrubs are noise control. I don't need five solid contributors — I need one guy who controls outcomes and five warm bodies who stay out of his way. Giannis got me to third place on roster scraps. Imagine if I'd had actual depth behind him instead of waiver wire rejects.
That maintenance day whisper is real and it's haunting me. High-usage scorer, 1-seed program — that's either Boozer or someone in that tier. I'm not backing off my read until I see something concrete, but it's in my head and it's going to make me hesitate at the wrong moment if I'm not careful.
Top valuations
Cameron Boozer $65AJ Dybantsa $44Darius Acuff Jr. $38Labaron Philon $30Graham Ike $28
D
Derek Fontaine
31 · High school PE teacher and assistant JV basketball coach · 0 yrs fantasy
Knowledgecasual
Strategybalanced
SourcesESPN, SportsCenter, ESPN App
Risk
Posts seen23
Background
Grew up in Memphis, Tennessee, middle kid in a household where both parents worked in healthcare — mom as a dental hygienist, dad as an X-ray tech. Practical people who watched sports for fun, never for analysis. Derek played rec ball through high school but was never quite varsity-caliber, so he became an enthusiastic spectator instead. Studied kinesiology at University of Memphis, graduated without a clear plan, drifted into teaching. He makes decisions by feel first, then skims conventional wisdom to make sure he isn't obviously wrong. Never digs deep — reads the headline, glances at the box score, moves on.
Philosophy
"I just try to get a couple guys I know are actually good and not overthink the rest. I've watched people go crazy trying to win the draft and then their team falls apart anyway. Get the names you trust and fill in the gaps as you go."
Going into the draft
Boozer and Dybantsa are the two names I keep hearing everywhere — ESPN, the mock results, the tier lists, all of it. They're clearly the top guys. But I watched Marcus go forty percent on one player and spend the rest of the draft scraping the bottom. I'm not doing that. Whatever Boozer costs in the room, I'm walking away before I hit a third of my budget.
One-seeds are safer bets. They've got better paths through the bracket, better teams around them, and in a tournament where one loss ends your whole investment, I'd rather have a guy who's expected to play five or six games than some six-seed who might torch it in round one and go home.
Dybantsa's numbers are wild — twenty-five points a game is not normal — but BYU is a six-seed. That's a real risk. I've been reading the hype all week and part of me wants to go get him, but then I think about how a team with less talent around him could lose to a three-seed and I've got a $40 doorstop.
Raised in Naperville, Illinois by an actuary father and a biostatistics professor mother. Family debates centered on Bayesian inference and expected value, never sports. Studied applied mathematics at Carnegie Mellon, then joined a Chicago quant desk at 23. A early career loss following a colleague's gut-feel stock tip cost him three months of savings and permanently hardened his conviction: sentiment is noise, models are signal. He has since applied that framework to every decision domain in his life.
Philosophy
"The draft is a portfolio allocation problem under uncertainty. Maximize projected value per dollar spent, overweight high-variance depth plays when the auction market misprice them due to name aversion, and never let familiarity substitute for the numbers."
Going into the draft
The market is systematically mispricing AJ Dybantsa. His 25.3 PPG is generating $47-58 consensus clearing prices in my sims, but ETP of 51.7 reflects what a 6-seed actually produces across expected tournament games. That is a 60-90% premium over model value driven entirely by a PPG narrative that ignores seed-weighted game count. I will not bid above $22. Let everyone else light their budgets on fire.
Thomas Haugh and Brayden Burries are the clearest value asymmetries in this pool. ETP 70.0 and 69.7 respectively, both on 1-seeds, both priced low because neither has a recognizable name in this room. That is the same arbitrage I hit in year three with the unknown center. Name-recognition discount plus deep-team upside is the single most repeatable edge in auction drafts.
Duke concentration is the trap this year. Boozer is the consensus darling and will clear above my model ceiling. Evans and Knueppel offer legitimate ETP from the same 1-seed run but I will not stack Duke — one round-one exit eliminates the entire position. One player per team is a hard constraint, not a preference.
SourcesESPN highlights, TNT broadcasts, friends' opinions at the bar
Risk
Posts seen19
Background
Grew up in Modesto, California in a loud household where his dad and uncles argued about games over beer and bet on players by feel and reputation. Never went to college — started doing yard work at 18 and eventually built his own crew. Business decisions have always come from pattern recognition and handshake instinct, not spreadsheets. When he took a chance on a young foreman who just looked like a leader and it paid off, it locked in his worldview: you watch someone long enough, you know what they are.
Philosophy
"I go by what I see. If a guy is out there looking like a problem every time I watch, I'm spending money on him. Stats can say whatever they want — I trust my eyes. I've played long enough to know you gotta swing big on the guys you believe in."
Going into the draft
Cameron Boozer is my guy and I don't care what I have to pay. Duke 1-seed, consensus number one, I've watched this kid and he's a problem every single time he touches the ball. That's my anchor and I'm not flinching when bidding opens.
Dybantsa dropping 25-plus as a 6-seed — I've been hearing his name everywhere this week and if BYU gets on a run that guy is going to go absolutely crazy. But I'm not paying top dollar for a 6-seed. If the price gets silly I'm stepping back.
Stack the 1-seeds. More games means more points. I want guys who are still playing in week two and three. A 4-seed guy averaging 22 sounds nice until they catch a hot team in round one and he's done scoring forever.
52 · High School Athletic Director · 17 yrs fantasy
Knowledgenone
Strategybalanced
SourcesESPN SportsCenter, sports talk radio, watching live games, coaching staff informal network, ESPN fantasy app, local sports newspaper
Risk
Posts seen21
Background
Raised in a Polish-American household in Cleveland where Sunday afternoons meant folding chairs around the television and her father's running commentary on every play. Her dad evaluated people by how they carried themselves — 'watch his footwork, not the scoreboard.' Marlene played varsity volleyball and softball, trained her eye for talent early. Spent twenty-five years as an athletic director scouting student athletes and writing scholarship recommendations. A bad stock tip from a colleague in 2003 that wiped out her vacation fund soured her permanently on data-driven decisions. She trusts what she sees with her own eyes and reads body language the way other people read box scores.
Philosophy
"I watch the games. I can tell when a guy has the look in his eyes or when he is playing scared. Stats do not show you body language or whether a receiver actually wants the ball in traffic. I would rather draft someone I have seen compete under pressure than trust a spreadsheet some kid made."
Going into the draft
I watched six hours of Cameron Boozer tape and that kid has footwork I have not seen from a freshman in twenty-five years of watching athletes. My dad always said watch the footwork, not the scoreboard. Boozer is my anchor tonight and I do not care what the simulation says his price should be.
I am not blowing my whole budget chasing names in the first fifteen minutes. I watched everyone else do exactly that in 2021 and I walked out of that draft with a balanced roster and a trophy. Let the room get emotional. I will stay patient and pick up value while they empty their wallets.
You cannot stack players from the same team. One bad call, one hot-shooting underdog, and two of your six players go dark at the same time. I am spreading across at least four teams minimum. That is not negotiable.
Top valuations
Cameron Boozer $56Thomas Haugh $40Darius Acuff Jr. $36Brayden Burries $32Graham Ike $29
What's next
More personas (50+) and more social sims (10+) for tighter confidence intervals
Full OASIS integration — follower graphs, reposting, algorithmic feeds
Real-time seed data — live odds, injury reports before each draft
Post-draft calibration — compare predictions to actual prices, tune the model
Fantasy football and basketball modules using the same persona + auction architecture