F1 Fantasy 2026 Predictions, Optimizer & Tips
Free, data-driven predictions for every 2026 F1 Fantasy round — machine-learning driver and constructor projections, a lineup optimizer, transfer planner, and weekly tips, with a public accuracy record. Updated every race weekend.
Pit Stop Performance — All Teams Stationary (wheels-up) times • click to collapse
No pit stop data yet — comes online after the first race of the season.
Lineup Optimizer
Find the best F1 Fantasy lineups within your budget, or get transfer recommendations for your current team.
How does the Lineup Optimizer work?
Checks all 1.4 million possible team combinations to find the best lineup within your budget. Uses ML predictions for qualifying and race positions.
- Lock picks: Left-click a driver/constructor card to lock them in your lineup
- Exclude picks: Right-click to exclude a player from consideration
- Chips: Select your active chip to see optimized lineups with boost applied
Lock / Exclude Picks
Left-click to lock (force into lineup). Right-click to exclude (remove from consideration). Constructors shown in ALL CAPS.
Data Analysis
No FP analysis data available. Run pipeline/10_fp_analysis.py first.
No post-race analysis data available.
2026 Season
Race Calendar & Results
Championship Standings
Official F1 World Championship points based on race and sprint results.
Fantasy Points Standings
Driver Price Tracker
Constructor Price Tracker
Head-to-Head Matchup
Compare any two drivers side-by-side. See who to pick based on predictions, value, and historical performance.
Model Accuracy
Track how well our predictions match reality. Updated after each race weekend.
Loading accuracy data...
Race Deep Dive
Detailed post-race telemetry analysis with fuel-corrected pace, sector breakdowns, and tyre degradation.
Select a round to view deep dive analysis...
Latest Videos
Race weekend analysis, predictions, and commentary from the BoxBox F1 Fantasy YouTube channel.
Subscribe on YouTubeLoading videos...
Race Weekend Articles
Analysis, insights, and commentary from each race weekend.
Loading articles...
What's New
Model and feature updates — the meaningful changes behind your fantasy picks.
Loading changelog...
Support BoxBox
This site is a passion project that I pour countless hours into. I love F1 and F1 Fantasy and I want to share that passion with everyone. I try to provide as much value as possible for free. Everything on here is free. Eventually I might add a subscription tier for advanced features, but I'll be very clear about what's free vs paid when that comes.
For now: tips just help cover hosting and keep the coffee flowing during those 2am coding sessions. If this site or any of my content has been helpful, please consider buying me a coffee. Every bit of support is much appreciated and goes directly into improving this site and maintaining it.
Many thanks,
Quintin
How It Works
BoxBoxF1Fantasy turns raw F1 data into fantasy point projections through a machine learning pipeline. Here's the high-level picture of what happens between a session ending and the numbers appearing on this site.
1. Data
We pull telemetry, results, and historical data from public F1 data sources covering 2020 onward. New data ingests after each session — practice, qualifying, sprint, and race.
2. Features
Each driver and constructor gets a feature vector blending historical form, recent results, track-specific patterns, team and driver skill ratings, and free practice pace when available.
3. ML Predictions
Gradient-boosted ranking models predict qualifying and race finish positions. The race model takes the qualifying prediction as input. Predictions sharpen as the weekend progresses and more data becomes available.
4. Monte Carlo
10,000 simulations per round add calibrated noise to the predictions and sample DNFs, overtakes, fastest laps, and Driver of the Day. This produces the 90% confidence intervals you see on driver cards.
5. Fantasy Points
Simulated outcomes are converted to fantasy points using the official 2026 F1 Fantasy scoring rules — including chips, sprint scoring, constructor pit stop bonuses, and the qualifying teamwork bonus.
6. Post-Race Analysis
After each race the deep-dive pipeline processes full lap data — fuel-corrected pace, tyre stints, sector analysis, real wheels-up pit stop times, and overtake counts — viewable in the Race Deep Dive tab.
Known Limitations & Quirks
Some things you might notice that look odd at first — they're working as intended.
- Wide confidence intervals are honest, not noisy. A 90% CI of "8 — 53 pts" reflects real F1 volatility (DNFs, safety cars, weather, strategy). When the model is genuinely uncertain, it says so rather than pretending otherwise. A narrow CI on a leader and a wide CI on a midfielder is usually correct, not a bug.
- Predictions shift across the weekend. Pre-FP predictions rely on historical patterns only (~65% confidence). Once practice data lands, telemetry sharpens the picture (~90% confidence). Qualifying then locks in the grid and tightens the race forecast further. If a driver moves up or down between Friday and Sunday, that's the model integrating new evidence.
- Sprint weekends have less data to work with. Sprint weekends compress practice into a single FP1 session before sprint qualifying. Less telemetry means slightly lower confidence and wider intervals than a regular weekend.
- The Multi-Week Planner's future rounds use heuristics, not real ML. Only the current round has full model predictions. Future rounds (2-5 weeks ahead) project using recent form weighted by track similarity to past circuits. Treat them as directional, not precise.
- New teams lean conservative. Cadillac has no 2020-2025 history to learn from, so their priors fall back toward grid averages. As 2026 results accumulate, their predictions will sharpen and reflect their actual pace.
- 2026 has fewer historical signals than usual. New regulations (active aero, redesigned ground effect, new power units) mean older data is partially out-of-distribution. The model weights 2026 races more heavily than older seasons, but early-season predictions carry more uncertainty than they would mid-season under stable rules.
- The calendar skips R4 and R5. Bahrain (R4) and Saudi Arabia (R5) were cancelled from the 2026 calendar but their round numbers are preserved. Miami is R6, Canada is R7, etc. — Round 4 and 5 just aren't shown.
- Constructor scoring never includes driver boost multipliers. Per official F1 Fantasy rules, the 2x/3x boosts (Autopilot, Final Fix, 3x Boost) only apply to driver scoring, never constructor scoring. Optimizer and predictions follow this rule.
- DNF probability blends history and current season. A driver's reliability rating is a mix of their rolling 5-race DNF rate and their team's 2026 reliability so far. Small sample sizes early in the year can produce DNF estimates that look high; they self-correct as more races run.
- Pit stop points may update a day after the race. Constructor pit stop scoring uses real wheels-up stationary times from public data feeds. Those feeds occasionally lag 24-48h post-race. If pit stop points look low immediately after the race, check back the next day — the numbers refresh automatically.
Frequently Asked Questions
How does F1 Fantasy scoring work?
Drivers score points for their qualifying and race finishing positions, positions gained, overtakes (+1 each), and fastest-lap and Driver of the Day bonuses, minus a penalty if they don't finish (DNF). Sprint weekends add sprint qualifying and sprint race points. Constructors score both of their drivers' points plus a qualifying teamwork bonus and pit-stop points. BoxBoxF1Fantasy projects every one of these for the upcoming round.
Is it better to pick good drivers or constructors in F1 Fantasy?
You need both, but don't overlook constructors: they score both of their drivers' points and add pit-stop points and a qualifying teamwork bonus that individual drivers don't get. A strong constructor is often better value than a third premium driver. Our Constructors tab shows the full breakdown so you can compare.
How do you win F1 Fantasy?
Consistency and value win leagues, not chasing one big week. Build around points-per-million value picks, plan transfers a few rounds ahead, time your chips for weekends that suit them, and captain a driver with high, reliable upside. BoxBox's Optimizer, Transfer Advisor and Multi-Week Planner are built to do exactly that.
How accurate are BoxBox's F1 Fantasy predictions?
We publish our track record: the Accuracy tab shows our prediction error and confidence-interval coverage for every completed round — including the misses. Predictions come from machine-learning models plus a 10,000-run Monte Carlo simulation, and they sharpen as practice and qualifying data arrive through the weekend.
Is BoxBoxF1Fantasy free?
Yes — all predictions and tools are completely free, with no login required.
Contact
Questions, bug reports, feature requests, or just want to argue about predictions? Reach out.
Disclaimer
These predictions are for entertainment purposes only. F1 races are inherently unpredictable — weather, crashes, strategy calls, and reliability can dramatically alter outcomes. Use these predictions as one input for your fantasy decisions, not the only one.