- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Coaching Changes
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
AI NFL Playoff Picks Betting Thread Chatgpt vs Grok
Posted on 1/10/26 at 11:37 am
Posted on 1/10/26 at 11:37 am
I created a prompt that gets some solid results
I started the season with Chatgpt which went well first few weeks then nose dived. Started using grok and it was fire
Here is Chatgpt
Here is grok
I started the season with Chatgpt which went well first few weeks then nose dived. Started using grok and it was fire
Here is Chatgpt
quote:
FINAL TOP 3 ATS (CURRENT NUMBERS)
Rank
Pick
ATS Win %
Confidence
1
Patriots -3.5
59.4%
????½
2
Rams -10
58.1%
????
3
49ers +5.5
56.9%
???½
Here is grok
quote:
1. San Francisco 49ers +4.5 at Philadelphia Eagles (Total: 44.5 | Sun 4:30 p.m. ET)
Modeled ATS Win %: 62.4% (95% CI: 62.2% - 62.5%)
Edge: ~10.0%
Key Math & Rationale: Implied prob ˜52.4% Eagles cover (z=4.5/v44.5˜0.67, F(0.67)˜0.748 adjusted for push), but FPI 57% Eagles win tunes µ_model=2.4 (s=6.67). d=(2.4-4.5)/2=-1.05 shifts ?_Eagles=24.5-1.05=23.45, ?_49ers=20+1.05=21.05 (mean=2.4). Sims yield P(Eagles margin <4.5)=62.4%, CI width ~0.15% from n=500k. Edge from public 75% on Eagles (overvaluing home DVOA +2.1% vs. SF pass rush EPA +0.14 neutral-site); low total tightens variance (s=6.67) for 60%+ historical dog hit in playoff rematches (p<0.01 binomial).
2. Los Angeles Chargers +3.5 at New England Patriots (Total: 46.5 | Sun 8 p.m. ET)
Modeled ATS Win %: 62.2% (95% CI: 62.1% - 62.3%)
Edge: ~9.8%
Key Math & Rationale: z=3.5/v46.5˜0.51 implies ~58% Pats cover, but FPI 54% Pats win gives µ_model=1.4 (s=6.82). d=(1.4-3.5)/2=-1.05 ? ?_Pats=25-1.05=23.95, ?_Chargers=21.5+1.05=22.55 (mean=1.4). Sims: P(Pats margin <3.5)=62.2%, precise CI. Edge stems from market overreaction to Pats home (public 68% on NE) ignoring Chargers' road EPA +0.12; mid-total variance (s=6.82) favors 59% dog cover in cold spots (p=0.005, 28°F forecast minimal impact).
3. Jacksonville Jaguars +1.5 vs. Buffalo Bills (Total: 52.5 | Sun 1 p.m. ET)
Modeled ATS Win %: 59.8% (95% CI: 59.6% - 60.0%)
Edge: ~7.4%
Key Math & Rationale: Line z=1.5/v52.5˜0.21 implies ~55% Bills cover, but FPI 51% Bills win (Jags by 0.3) yields µ_model=-0.3 (s=7.25). d=(-0.3-1.5)/2=-0.9 ? ?_Bills=27-0.9=26.1, ?_Jags=25.5+0.9=26.4 (mean=-0.3). Sims: P(Bills margin <1.5)=59.8%, CI width 0.2%. This is a sharp fade—public 70% on Bills masks Jags' home DVOA +1.8%; high total amplifies variance (s=7.25) for 58% underdog hit in dome playoffs (p=0.02).
#1 Best Odds Pick: San Francisco 49ers +4.5
Highest edge (10.0%) from undervalued SF metrics—sims 62.4%, Kelly ~0.09. Public trap projects 60%+ real hit. Skipped slimmer like Panthers +10.5 (57%). Re-sim for weather/injuries if lines shift.901cca9d23a3
6
Posted on 1/10/26 at 11:52 am to SDVTiger
I’ve been wondering when this would become prevalent in sports gambling. Obviously not 100%c but even a 55% accuracy rate could net some pretty nice profits.
Posted on 1/10/26 at 11:57 am to 9Fiddy
quote:
Obviously not 100%c but even a 55% accuracy rate could net some pretty nice profits.
You have to be 60% to break even in sports gambling.
Posted on 1/10/26 at 12:07 pm to SDVTiger
Did Grok return deepfake nudes of Trevor Lawrence?
Posted on 1/10/26 at 1:18 pm to 9Fiddy
I think grok was 18 and 10 the last i checked
Last win was Steelers over ravens
Its had Ole Miss +3.5
Last win was Steelers over ravens
Its had Ole Miss +3.5
Posted on 1/10/26 at 1:24 pm to SDVTiger
quote:
Its had Ole Miss +3.5
I am not Groot. I am Grok.
Posted on 1/10/26 at 4:06 pm to BamaCoaster
quote:
You have to be 60% to break even in sports gambling
52.4% to break even assuming the juice is 10%. If you are hitting at 60% with a large sample, you are world class.
Posted on 1/10/26 at 6:02 pm to 9Fiddy
James Carville went 6-0 last week.
Posted on 1/10/26 at 6:14 pm to BamaCoaster
quote:
You have to be 60% to break even in sports gambling.
60% makes you very profitablePosted on 1/10/26 at 6:19 pm to SDVTiger
I’m on
Panthers +10
Bears +1
Jags +1
Eagles -3.5
Chargers +3.5
Steelers +3
Panthers +10
Bears +1
Jags +1
Eagles -3.5
Chargers +3.5
Steelers +3
Posted on 1/10/26 at 6:29 pm to MrFreakinMiyagi
Wrong thread
This post was edited on 1/10/26 at 6:31 pm
Posted on 1/10/26 at 9:07 pm to PeteRose
quote:
PeteRose
quote:This guy knows a thing or 2 about sports betting.
52.4% to break even assuming the juice is 10%. If you are hitting at 60% with a large sample, you are world class.
Posted on 1/11/26 at 2:22 am to MrFreakinMiyagi
quote:
I’m on Panthers +10 Bears +1 Jags +1 Eagles -3.5 Chargers +3.5 Steelers +3
Does MSB not keep a betting thread this season. Disappoint
Posted on 1/11/26 at 7:27 am to MrFreakinMiyagi
quote:
I’m on
Panthers +10
Bears +1
Jags +1
Eagles -3.5
Chargers +3.5
Steelers +3
I like all.of these. Agreed with panthers +10
AI gave the reason it added them in the top 3. It was sound but thats a lot of points in a playoff game
Posted on 1/12/26 at 7:39 am to SDVTiger
Chatgpt 2-1
Grok 1-2
Best bet for Grok and Chatgpt were winners
For tonights game both are picking the steelers
Chatgpt
Grok
Grok 1-2
Best bet for Grok and Chatgpt were winners
For tonights game both are picking the steelers
Chatgpt
quote:
FINAL PLAY (TONIGHT)
?? Steelers +3
ATS Win Probability: 58.6%
Confidence Rating: 4 / 5
Best Use: Straight bet or teaser leg
CLV Note: If you can find +3.5, this jumps above 61%
Grok
quote:
The Pick: Pittsburgh Steelers +3
Strong playoff prep winner—sims 64.8%, EV ~12% per unit (Kelly ~0.11). Cold + Watt + Tomlin home dog magic = backdoor cover potential. Fade the road chalk; Steelers keep it close for momentum. Re-sim post-kickoff if Jordan/Austin tweaks. Congrats on the SF hit—let's stack another! ?? #NFLPlayoffs #ATS
Posted on 1/12/26 at 10:27 am to SDVTiger
What's the prompt?
Posted on 1/12/26 at 10:28 am to SDVTiger
I just asked Chatgpt about tonights game. Spit out Steelers +3.5
Posted on 1/12/26 at 12:18 pm to Zendog
quote:
Act as a professional sports betting analyst with a strong background in applied mathematics, probability, and NFL advanced metrics. Each week, I will provide you with 5 NFL matchups. For each matchup, break down the game using:
Statistical models (DVOA, EPA per play, success rate, yards per play, QB efficiency, red zone %, turnover differential).
Market factors (line movement, consensus spreads, sharp money vs public money, historical ATS trends, injury reports, weather, rest days, travel factors).
Mathematical projections (expected score distributions, Poisson models, Monte Carlo simulations, implied probabilities vs bookmaker odds).
Output for each game should include:
Predicted final score range (with probabilities).
ATS pick (team + or - spread) with confidence %.
Any totals/over-under lean with supporting math.
A summary of why the edge exists vs the market.
Give me the best 3 ATS picks of the week with your quantified confidence rankings (from 1–5).
Keep analysis professional, math-driven, and focused on beating the closing line value (CLV).
Posted on 1/17/26 at 8:50 am to SDVTiger
Chatgpt 2-2
Grok 1-3
Heres todays picks
Grok
Chatgpt
Best of luck with your picks today
Grok 1-3
Heres todays picks
Grok
quote:
. San Francisco 49ers +7 at Seattle Seahawks (Total: 45.5 | Sat 4:30 p.m. ET)
Modeled ATS Win %: 68.5% (95% CI: 68.3% - 68.7%)
2. Buffalo Bills +1.5 at Denver Broncos (Total: 45.5 | Sat 8 p.m. ET)
Modeled ATS Win %: 58.5% (95% CI: 58.3% - 58.7%)
Edge: ~6.1%
3. Houston Texans +3 at New England Patriots (Total: 41 | Sun 3 p.m. ET)
Modeled ATS Win %: 60.2% (95% CI: 60.0% - 60.4%)
Edge: ~8.2%
#1 Best Odds Pick: San Francisco 49ers +7 (18.9% edge—public trap, Kelly ~0.13). #2: Buffalo Bills +1.5 (6.1% edge—cold fade). Third swaps to Texans for stronger edge. Skipped Rams +3.5 at Bears (58.1%, slimmer in brutal cold). Deep dive rebounds us—re-sim post-inactives.
Chatgpt
quote:
1) Rams @ Bears — Bears +4.5
Cover probability: 52.2%
Confidence: 3/5
Why it’s #1: Chicago’s wind/snow/cold increases drive volatility and suppresses explosive efficiency, which mathematically helps the underdog cover mid spreads. ?
NBC Sports
2) Texans @ Patriots — Patriots -3
Cover probability: 52.2%
Confidence: 3/5
Why it’s #2: Collins being out is a real downgrade to Houston’s passing efficiency and 3rd-down conversion ceiling. ?
Note: at -3, pushes matter in real scoring (my continuous sim doesn’t “push” on exactly 3 the way NFL scoring does). If you can get -2.5, this improves meaningfully.
Reuters +1
3) 49ers @ Seahawks — Seahawks -7
Cover probability: 51.4%
Confidence: 2.5/5
Why it’s #3: SF’s injury stack (Kittle IR + Warner out) shifts the distribution toward Seattle covering. ?
But at -7, you’re sitting on a key number, so the edge is thinner unless you find -6.5.
Best of luck with your picks today
Page 1 of 1
Back to top
Follow TigerDroppings for LSU Football News