SharperGamblerSHARPERGAMBLER
Round 16

Bahrain Grand Prix

Kuala Lumpur · Bahrain
Lap -Distance - Format conventional
Next: Practice 1
--:--:--

Weekend schedule

updated
Practice 1
scheduled
Sprint Qualifying
scheduled
Sprint
scheduled
Qualifying
scheduled
Race
scheduled

Weather

updated

Forecast unavailable at last regeneration.

Predictions | every version frozen & scored

updated

Qualifying prediction

Showing LIVE, recomputed from current data · stage Pre-weekend (season floor)
this IS the season-average floor, the permanent benchmark, kept and scored every weekend
SPRINT WEEKEND: the practice blend cannot run. There is no FP2 or FP3 before qualifying, and FP1 is not a validated session for this model.
The official qualifying prediction for this round is therefore the season-average floor, exactly as it was before the blend was promoted. This is a clean fallback, not a degraded blend.
#DriverPole%Top3%Top10%
1RUS2559100
2ANT1952100
3HAM154599
4NOR144499
5LEC123999
6VER93399
7PIA52297
8HAD0482
9LAW0154
10GAS0051
11LIN0042
12COL0024
13HUL0022
14BOR0014
15TSU005
16SAI004
17BEA004
18OCO002
19ALB001
20ALO000
21PER000
22BOT000
23STR000

No frozen version for this stage yet; the live figures above will be frozen when this stage locks.

Official version: after FP3, before qualifying (not locked yet).

Season trend

Us vs the floor FOR THAT SAME WEEKEND, newest first. Green = we beat the floor that weekend. Places-off and Brier: lower better; rank-corr: higher better.

WkndPlaces offRank corrBrierOur version
R153.09 / 4.000.75 / 0.720.07 / 0.10After FP3 (official)
R142.05 / 2.000.95 / 0.930.07 / 0.07After FP3 (official)
R132.68 / 2.960.88 / 0.830.08 / 0.10After FP3 (official)
R121.81 / 1.810.94 / 0.940.06 / 0.06Pre-weekend (season floor)
R111.46 / 2.090.96 / 0.920.06 / 0.07After FP3 (official)

The floor is a METHOD, not a fixed number. It is each driver's season-average position, re-scored against every race, so it lands somewhere different every weekend depending on how chaotic that race was. Across 2026 it has scored anywhere from 2.27 to 7.36 places off. Only the within-weekend comparison in this table is valid; comparing this weekend's result against another weekend's floor compares two different races and means very little.

Race prediction

Showing LIVE, recomputed from current data · stage Floor (season grid)
this IS the season-average floor, the permanent benchmark, kept and scored every weekend
PLACEHOLDER, not a model. Finishing position is predicted from grid slot only. No driver, team, car pace or race pace is used, so it structurally cannot reorder the front of the grid.
A real race model is in development; this is the baseline it must beat.
#DriverWin%Pod%Pts%DNF%
1RUS34658218
2ANT22578416
3HAM16488216
4NOR13428117
5LEC6267718
6VER4217518
7PIA2137116
8HAD186516
9LAW176216
10GAS055417
11LIN034917
12COL024216
13HUL013418
14TSU012618
15BOR002416
16SAI001819
17BEA001516
18ALB001317
19OCO001217
20ALO001217
21BOT00918
22PER00818
23STR00518

No frozen version for this stage yet; the live figures above will be frozen when this stage locks.

Official version: after qualifying is analysed (official grid) (not locked yet).

Season trend

Us vs the floor FOR THAT SAME WEEKEND, newest first. Green = we beat the floor that weekend. Places-off and Brier: lower better; rank-corr: higher better.

WkndPlaces offRank corrBrierOur version
R155.09 / 4.590.47 / 0.630.14 / 0.15After qualifying (official)
R142.30 / 3.230.77 / 0.700.08 / 0.09After qualifying (official)
R132.91 / 2.960.71 / 0.790.10 / 0.08After qualifying (official)
R123.00 / 3.190.74 / 0.710.10 / 0.11After qualifying (official)
R112.36 / 3.360.80 / 0.750.08 / 0.09After qualifying (official)

The floor is a METHOD, not a fixed number. It is each driver's season-average position, re-scored against every race, so it lands somewhere different every weekend depending on how chaotic that race was. Across 2026 it has scored anywhere from 2.27 to 7.36 places off. Only the within-weekend comparison in this table is valid; comparing this weekend's result against another weekend's floor compares two different races and means very little.

Last completed round | Azerbaijan Grand Prix

updated

The next weekend has not started yet, so these are the results from Round 15, Azerbaijan Grand Prix, the most recent completed round. This section is replaced by the live weekend automatically as soon as the first session of the next round is collected.

Qualifying

PosDriverBestQ1Q2Q3
1George Russell1:42.5261:43.6151:43.4621:42.526
2Charles Leclerc1:43.3631:44.3601:43.7801:43.363
3Oscar Piastri1:43.3641:45.0141:43.8141:43.364
4Isack Hadjar1:43.5001:44.1611:43.8801:43.500
5Lando Norris1:43.6721:44.5711:44.0201:43.672
6Lewis Hamilton1:43.8581:44.2601:44.0371:43.858
7Pierre Gasly1:44.0471:44.4891:44.1061:44.047
8Max Verstappen1:44.0811:44.0411:43.7061:44.081
9Carlos Sainz1:44.5661:45.1041:44.6291:44.566
10Franco Colapinto1:44.9631:45.1061:44.6831:44.963
11Oliver Bearman1:44.7751:45.2281:44.775-
12Liam Lawson1:44.8601:45.5351:44.860-
13Alexander Albon1:45.0011:45.0311:45.001-
14Esteban Ocon1:45.0161:45.0391:45.016-
15Arvid Lindblad1:45.1061:45.3811:45.106-
16Kimi Antonelli1:45.5041:45.504--
17Gabriel Bortoleto1:45.7991:45.799--
18Nico Hulkenberg1:45.9201:45.920--
19Fernando Alonso1:46.5931:46.593--
20Sergio Perez1:46.6581:46.658--
21Lance Stroll1:47.3371:47.337--
22Valtteri Bottas1:48.2901:48.290--

Race

Average cleaned lap time per compound: lap 1, pit in/out laps, safety-car / VSC / red-flag laps, deleted and inaccurate laps all excluded. Multiple stints on the same compound are shown separately, each badged with its stint number. Only compounds actually used this session get a column. Fewer than 3 clean laps in a stint = a dash, not a misleading number. softmediumhardinterwet

PosDriverBestSoftMedium
1George Russell1:44.91631:45.35811:47.712
2Max Verstappen1:44.99331:45.37911:47.975
3Isack Hadjar1:45.61811:48.07531:46.152
4Charles Leclerc1:45.78411:48.27031:46.310
5Kimi Antonelli1:45.41331:46.43211:48.475
6Lewis Hamilton1:46.17011:48.46931:46.942
7Arvid Lindblad1:46.84931:47.52911:49.616
8Esteban Ocon1:47.06811:49.43931:47.542
9Oliver Bearman1:46.97811:49.23831:47.700
10Carlos Sainz1:46.73721:48.77041:47.61011:50.298
11Nico Hulkenberg1:46.85031:47.77011:49.907
12Liam Lawson1:46.90611:49.40431:47.719-
13Gabriel Bortoleto1:46.33631:47.79411:49.811
14Oscar Piastri1:46.60211:48.05731:47.298
15Sergio Perez1:47.36331:48.30311:50.056
16Valtteri Bottas1:48.85231:49.59111:51.989
17Franco ColapintoRetired1:48.484-11:49.098
18Pierre GaslyRetired1:47.822-11:48.646
19Lando NorrisRetired1:47.761-11:48.638
20Alexander AlbonRetired1:48.44311:49.471-
21Fernando AlonsoRetired1:50.96811:52.074-
22Lance StrollRetired1:51.72311:52.090-

Starting grid

updated

Awaiting qualifying. The starting grid is published after qualifying and after any penalties are applied, so there is nothing to show yet. This section stays empty rather than showing a predicted order, because a prediction is not a grid.

Results

updated

No completed sessions yet this weekend.

Track sector map

updated

The circuit layout for Kuala Lumpur will appear here after the next automatic data refresh (it is fetched once, without any telemetry download).

Grip vs energy | where the lap time went

updated
DO NOT compare TEAMS on the straight-line column. The grid is now narrower than the instrument. Across 2026 qualifying the whole field spans 0.340s on that column against a resolution of about 0.35s, and only 8 of 55 team pairings have non-overlapping confidence intervals, with the top six indistinguishable. The between-team ranking supports no team-versus-team claim and should not be cited. Levels are flattered too: sessions discarded by the offsetting check are strongly biased toward FAST circuits (correlation +0.79), so every team is measured mostly on slower rounds than the calendar average.
USE THIS INSTEAD: the WITHIN-TEAM contrast, mechanical minus straight-line. The 0.35s limit does not apply to it. That figure was the spread of the same component between two DIFFERENT cars; this compares two components of the same car, same session, same reference lap, same driver, holding constant everything that makes a between-car comparison noisy. 10 of 11 teams are separable on it, against 8 of 55 pairings between teams. A positive number means the car loses more in the corners than on the straights.
Most grip-weak: Cadillac +3.14, Aston Martin +2.75. Most balanced: Mercedes +0.13. Ferrari is the only NEGATIVE (-0.17), losing more on the straights than in the corners, corroborated independently by the superclip metric, but its interval [-0.36, +0.02] includes zero, so it is PARKED, UNDERPOWERED, LEANING, neither confirmed nor refuted. See reports/straightline_ranking_2026.md.

Monza qualifying. Reference lap GAS 81.786s. Across this session 91% of all time lost was mechanical, 9% was lost on straights, and 0% is unattributed : full throttle through faster curves, where grip and power limit the car at once. Mechanical plus energy is therefore not the whole story.

For some teams the two cars' split offsets, so read the gap and not the breakdown. This weekend that applies to Aston Martin, Mercedes: their mechanical and straight-line figures swing against each other by more than the two cars are actually apart. The total gap is still right.
Two things can cause it and we cannot yet tell them apart: a genuine setup difference between the cars (different wing level: quicker on the straights, slower in the corners, similar lap time), or a limitation of the split. It is judged per session, not per team, because instability turns out to follow the circuit far more than the team.
TeamGapMechanicalStraight-lineUnattributedIn words
Mercedes+0.078+0.151-0.064-0.0090.08s off: the two cars' split offsets, so read the gap, not the breakdown
Alpine+0.217+0.154+0.054+0.0090.22s off: 0.15s mechanical, 0.05s straight-line
Ferrari+0.221+0.297-0.059-0.0160.22s off: 0.30s mechanical, -0.06s straight-line
McLaren+0.230+0.236+0.000-0.0060.23s off: 0.24s mechanical, 0.00s straight-line
Red Bull Racing+0.659+0.813-0.145-0.0080.66s off: 0.81s mechanical, -0.15s straight-line
Audi+0.862+0.955-0.082-0.0110.86s off: 0.96s mechanical, -0.08s straight-line
Racing Bulls+1.234+1.517-0.278-0.0041.23s off: 1.52s mechanical, -0.28s straight-line
Haas F1 Team+1.319+1.330-0.014+0.0031.32s off: 1.33s mechanical, -0.01s straight-line
Williams+2.118+1.064+0.997+0.0582.12s off: 1.06s mechanical, 1.00s straight-line
Cadillac+2.694+2.632+0.050+0.0122.69s off: 2.63s mechanical, 0.05s straight-line
Aston Martin+3.400+2.700+0.697+0.0023.40s off: the two cars' split offsets, so read the gap, not the breakdown

Sections are classified by lateral load, not by curvature: a corner taken at 0.7g is not grip-limited when the car can pull 5g. The climbing-power term is removed using track altitude. See reports/grip_energy_split.md.

Track history |

updated

All computed from our data (0 races with lap/results on file). Small samples flagged.

0/0
pole → win
+nan
avg grid→finish change (±nan)
0%
historical DNF rate

Safety cars: no data

Pit strategy

Not enough clean pit cycles at this circuit yet to grade the undercut. It needs several races on file.

Position change

Not enough races at this circuit yet to measure how positions change here. It needs at least three on file.

Circuit character | the five dimensions

This circuit is not in the measured table yet, which happens on a first visit or a renamed venue. It appears once the round has race data on file.

Best current drivers here (avg finish)

DriverAvg finishRaces

News (last 7 days)

updated

Regulation first, because it can invalidate our data. Gossip and other series already stripped.

Power rankings

updated
Read the WEEKEND column as DESCRIPTIVE, not a forecast. It reports who has actually been quick this weekend on a like-for-like tyre. We have MEASURED that this does NOT beat season average as a predictor of the result, so do not treat it as a prediction. The SEASON column is our validated form measure.

Weekend column source: awaiting practice.

Drivers

DriverSeasonWeekendShift
RUS1-awaiting
ANT2-awaiting
HAM3-awaiting
LEC4-awaiting
NOR5-awaiting
PIA6-awaiting
VER7-awaiting
HAD8-awaiting
GAS9-awaiting
LAW10-awaiting
LIN11-awaiting
HUL12-awaiting
COL13-awaiting
BOR14-awaiting
BEA15-awaiting
TSU16-awaiting
OCO17-awaiting
SAI18-awaiting
ALB19-awaiting
ALO20-awaiting
PER21-awaiting
BOT22-awaiting
STR23-awaiting

Teams

TeamSeasonWeekendShift
Mercedes1-awaiting
Ferrari2-awaiting
McLaren3-awaiting
Red Bull Racing4-awaiting
Alpine5-awaiting
Audi6-awaiting
Racing Bulls7-awaiting
Haas F1 Team8-awaiting
Williams9-awaiting
Aston Martin10-awaiting
Cadillac11-awaiting

Season archive | 2026

updated

Every completed round stays here after the page rolls over to the next weekend. Frozen predictions are kept in the database and are never overwritten; the linked files open in this browser.

RoundEventCollectedPredictionsFiles
R15Azerbaijan Grand Prixrace complete4 qualifying / 5 race versions frozen-
R14Spanish Grand Prixrace complete4 qualifying / 5 race versions frozen-
R13Italian Grand Prixrace complete4 qualifying / 5 race versions frozen-
R12Dutch Grand Prixrace complete2 qualifying / 3 race versions frozenqualifying scorecard | race scorecard
R11Hungarian Grand Prixrace complete6 qualifying / 6 race versions frozenqualifying scorecard | race scorecard | race report | qualifying prediction
R10Belgian Grand Prixrace complete4 qualifying / 5 race versions frozen-
R09British Grand Prixrace complete2 qualifying / 3 race versions frozen-
R08Austrian Grand Prixrace complete4 qualifying / 5 race versions frozen-
R07Barcelona Grand Prixrace complete4 qualifying / 5 race versions frozen-
R06Monaco Grand Prixrace complete4 qualifying / 5 race versions frozen-
R05Canadian Grand Prixrace complete2 qualifying / 3 race versions frozen-
R04Miami Grand Prixrace complete2 qualifying / 3 race versions frozen-
R03Japanese Grand Prixrace complete4 qualifying / 5 race versions frozen-
R02Chinese Grand Prixrace complete2 qualifying / 3 race versions frozen-
R01Australian Grand Prixrace complete4 qualifying / 5 race versions frozen-