Daggy vs. the Market

Daggy, drawn as the saloon's robot bartender
Daggy
W-L Record
1-1
50.0%
Brier score0.277
VS
Kalshi, drawn as the saloon's cash register — the market price
Kalshi
W-L Record
1-1
50.0%
Brier score0.283
0 · perfect0.25 · coin flip
Daggy 0.277Kalshi 0.283
Lower Brier = better forecasts. 0 is a crystal ball, 0.25 is a coin flip.
2 settled games so far
Weeks 1–5

Week 1

NEatSEA
WED, SEP 9 · 8:20 PM ET · Week 1 · Final
Daggy picked
SEA
brier 0.15
Kalshi priced
SEA
brier 0.14
FINAL WINNER: SEA
SFatLA
THU, SEP 10 · 8:35 PM ET · Week 1 · Final
Daggy picked
LA
brier 0.40
Kalshi priced
LA
brier 0.42
FINAL WINNER: SF
CHIatCAR
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
CHI
59% win prob
Kalshi price
CHI
60¢ implied
TBatCIN
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
CIN
64% win prob
Kalshi price
CIN
65¢ implied
NOatDET
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
DET
74% win prob
Kalshi price
DET
75¢ implied
BUFatHOU
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
BUF
52% win prob
Kalshi price
BUF
52¢ implied
BALatIND
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
BAL
62% win prob
Kalshi price
BAL
61¢ implied
CLEatJAX
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
JAX
78% win prob
Kalshi price
JAX
79¢ implied
ARIatLAC
SUN, SEP 13 · 4:25 PM ET · Week 1
Daggy picks
LAC
81% win prob
Kalshi price
LAC
81¢ implied
MIAatLV
SUN, SEP 13 · 4:25 PM ET · Week 1
Daggy picks
LV
61% win prob
Kalshi price
LV
60¢ implied
GBatMIN
SUN, SEP 13 · 4:25 PM ET · Week 1
Daggy picks
MIN
54% win prob
Kalshi price
MIN
54¢ implied
DALatNYG
SUN, SEP 13 · 8:20 PM ET · Week 1
Daggy picks
DAL
58% win prob
Kalshi price
DAL
59¢ implied
WASatPHI
SUN, SEP 13 · 4:25 PM ET · Week 1
Daggy picks
PHI
68% win prob
Kalshi price
PHI
68¢ implied
ATLatPIT
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
PIT
61% win prob
Kalshi price
PIT
69¢ implied
NYJatTEN
SUN, SEP 13 · 1:00 PM ET · Week 1
Daggy picks
TEN
56% win prob
Kalshi price
TEN
52¢ implied
DENatKC
MON, SEP 14 · 8:15 PM ET · Week 1
Daggy picks
KC
56% win prob
Kalshi price
KC
57¢ implied

How This Whole Racket Works

1

The Two Players

Daggy is not one model. It's 8 agents forecasting every game — 4 of them wired to the Daggy statistical engine, the rest working from what they can reason out — combined into a single pick per game, before kickoff.

Kalshi isn't a rival forecaster. It's the live price real money is trading at, on a market where anyone can bet on the outcome. When we say “Daggy vs. Kalshi,” we mean model vs. market — not two people guessing.

2

The Tally (W-L)

The simplest measure there is: did the pick win or lose? Easy to read at a glance, and it's the headline number for a reason.

3

The Real Score (Brier)

The Brier score checks whether the confidence was earned, not just whether the pick happened to land. Two forecasters can both go 1-0 on a game and still score very differently:

Said 90%, team won
0.01
Said 55%, team won
0.20
Lower Brier score = better forecasts. 0 is a crystal ball, 0.25 is a coin flip.
4

Who's Actually Predicting

8 LLM agents predict every game — 4 wired to the Daggy engine, 4 running solo.

Claude Opus 5 · GPT-5.5 · Gemini 3.1 Pro · Kimi K2.5
A supervisor watches all 8 picks — a second set of eyes, not the deciding vote yet.

Leaderboard

Kimi K2.5
Augmented
1-1 (50.0%)
Brier 0.280
vs
Unaugmented
1-1 (50.0%)
Brier 0.256
Claude Opus 5
Augmented
1-1 (50.0%)
Brier 0.268
vs
Unaugmented
1-1 (50.0%)
Brier 0.283
GPT-5.5
Augmented
1-1 (50.0%)
Brier 0.279
vs
Unaugmented
1-1 (50.0%)
Brier 0.283
Gemini 3.1 Pro
Augmented
1-1 (50.0%)
Brier 0.293
vs
Unaugmented
1-1 (50.0%)
Brier 0.272
Daggy