Glossary

Plain-English definitions for the metrics, jargon, and concepts used across FASTLAP. The same definitions back the little (?) popovers next to column headers and control labels.

Rankings & Power Rankings

Overall Rankoverall-rank: Blends each category column (track history, track type, tire fits, dominance, pit crew, practice) into one ordered list for the next race. A lower Overall is a stronger model lean — Overall 1 is the model's top pick. Sort any individual column to see which category drives a driver's overall position.
Track Historytrack-history: Mean position a driver has run through races at this exact track over the recent seasons in our dataset. We rank by avg running position (not finish) because it filters out late-race wrecks and ill-timed cautions, which is a cleaner read on actual pace.
Track Typetrack-type: Same metric as Track History, but pooled across every track sharing the same surface/length profile. Useful when a driver has limited starts at the specific track but a long body of work at its siblings.
Left Tireleft-tire: NASCAR runs different compounds on the left and right sides depending on the track. This rank is a driver's avg running position at every race historically run on the same left-side compound as the upcoming race — a pace signal at similar grip levels.
Right Tireright-tire: Right-side mirror of Left Tire. The right side is the higher-loaded tire in most oval racing, so this rank often correlates more strongly with race pace at intermediate and short tracks.
Tire Combotire-combo: Filters history to races that ran the exact left/right combo of the upcoming race. The tightest tire comp, but also the smallest sample — read it alongside Left Tire and Right Tire rather than alone.
Dominancedominance: Sums each driver's laps led and fast laps run at this track, then ranks. Drivers near the top are the historical front-runners here. Note this is a raw total — a driver who has run more races at the track has a structural advantage in the count.
Pit Crewpit-crew: Average time the car spends stationary in the pit box across the season's four-tire stops (the comparable workload). Faster crews keep their driver from losing track position on green-flag cycles and gain spots on caution-flag stops.
Practicepractice-rank: Each driver's best 10-lap consecutive average from the most recent practice session (after their first 4 laps are dropped to fade out junk laps). The cleanest read on real long-run speed for the weekend before qualifying. If no practice has been held yet, this column reads "—".
Avg Running Positionavg-running-pos: NASCAR samples car positions throughout the race. The mean of those samples is a much steadier read on pace than the final finish, which can swing 15 spots on one bad pit stop or a late-race wreck. Most of our pace-based ranks key off avg running position for that reason.
Luck Deltaluck-delta: Finish minus avg running position. If a driver ran 5th but finished 22nd, their luck delta is +17 — they were faster than the result. Sustained positive deltas flag drivers due for positive variance; sustained negatives flag drivers whose finishes have overrated their pace.

Race Simulator

Win %win-pct: Of the 5,000 or 10,000 race simulations you ran, the share where this driver crossed the line first. Use it as your model's win-probability number — directly comparable to a sportsbook's implied win odds once you strip vig.
Top 5 %top5-pct: The model's probability of a top-5. Powers the floor side of DFS lineups and the top-5 sportsbook markets. A driver with a thin Win % but a fat Top 5 % is a stable ceiling/floor play rather than a tournament dart.
Top 10 %top10-pct: The model's probability of a top-10. The strongest signal for cash-game DFS lineups, which reward reliable median outcomes more than ceiling spikes.
Optimal %optimal-pct: After each sim race, we solve for the highest-DK-point lineup that fits the salary cap and roster rules. Optimal % is the share of those sim-optimal lineups that included this driver. High Optimal % at a low salary is the textbook leverage play — undervalued by the market relative to how often the perfect lineup needs them.
Mean DK Pointsmean-dk: Mean DK fantasy points across every iteration. Compare to a driver's DK salary to spot value — high Mean DK at a low salary beats the rest of the field per dollar. Tournament players should weight Win % and Top 5 % alongside Mean to avoid chalking up on safe-floor plays.
Dominatordominator: In DK NASCAR scoring, laps led and fast laps are worth real points and a few drivers usually soak up most of them. Tagging a driver as a Dominator tells the sim to bias laps-led and fast-laps allocation toward them, which lifts their ceiling. Mark 1–4 per race; tagging the whole field defeats the purpose.
Boostboost: A small additive tilt to a driver's simulated pace. Single up-chevron = mild like, double up = strong like; mirrors apply on the downside. Use it to encode an opinion the raw inputs don't capture — a driver who tested well, a crew chief change, a weather-suited setup. It will not overpower base ratings if you use it sparingly.
Chaos Levelchaos-level: A single dial that scales how often simulated races feature blown engines, wrecks, and pit-road mistakes. Restrained = clean races. Historical = matches season-average attrition. Maximum = expect wild superspeedway-style results, with mid-pack drivers landing in the top 5 more often.
Tire Weartire-wear: Light = fresh-tire restarts dominate, position shuffles less. Medium = baseline. Heavy = sustained long-run speed separates the field, and any driver with poor tire management loses ground each run. Heavy wear at intermediates is a meaningful lever for elite long-run cars.

DFS strategy

GPPgpp: Large-field DFS contests where most of the prize money goes to the top few finishers. GPP strategy weights ceiling and uniqueness — drivers and stacks the field is fading. Contrast with cash games, where consistency beats upside.
Chalkchalk: A driver or stack that a large share of DFS lineups are projected to use. Chalk isn't bad — chalky plays are usually chalky for a reason — but in GPPs, two chalk plays cap your differentiation if you don't pair them with a contrarian piece.
Leverageleverage: A driver the model rates highly but the public is fading (low projected ownership). Leverage plays don't need to win outright — they win you tournaments when they finish above their ownership, because few lineups had them.
Fadefade: When the brief or the model flags a fade, it's saying their market price or DK ownership overstates their actual chance. Fading chalk is the main way to get unique in GPPs; fading a busted setup is the main way to protect cash-game floors.

Betting & odds

Implied %implied-pct: Convert American odds to a percentage. +200 implies ≈33%, −150 implies ≈60%. Sportsbooks pad these — the sum across a field exceeds 100% because the book's margin (vig) is inside every line. Use vig-stripped Fair % when comparing market to model.
Vig-stripped (Fair %)vig-stripped: We scale every driver's implied probability so the field sums to 100% (the "no-vig" or "fair" market). This is the apples-to-apples number to compare with the model's Win %. If model Win % > Fair %, the book is selling the driver too cheap by your model.
Market Rankmarket-rank: Drivers ordered by current win odds, shortest price first. The market's ranking compresses a ton of information (sharp money, injuries, weather) into one number — disagreements between Market Rank and Power Rankings Overall are where the most interesting bets live.

Pit stops

In-Box Timein-box-time: Stopwatch starts when the car stops in the pit box and ends when it leaves. This is the cleanest read on the crew, because it excludes pit-road speed and entry/exit lines (which are the driver's job). Sort the analyzer by avg in-box to see who the elite crews are this season.
Total Stop Timetotal-stop-time: Whole-stop duration end to end. Reflects pit-road length (a track variable), driver pit-in/out execution, and the crew's in-box work combined. Useful for ranking how much net time a driver loses on a green-flag stop at this specific track.
Stop Strategystop-strategy: NASCAR strategy splits on cycle: 4 tires for max grip, 2 tires for track-position, fuel-only for stretch attempts, no service to gain spots. Avg in-box times only compare apples-to-apples within the same strategy, so the analyzer groups by stop type.

Practice analyzer

Best-N Lap Averagebest-n-avg: Selects the N fastest legal laps from the session and averages them. We default to a Best-10 read after dropping each driver's first 4 laps (junk warmup). Best-N tracks raw single-lap pace; pair it with Best-Avg-Consecutive for long-run drop-off.
Best Consecutive Averagebest-consec-avg: Scans the whole session for the fastest N laps in a row and averages them. This is the long-run pace read — a driver with a strong consecutive average kept the car under them while tires wore. Best for evaluating race-trim setups.
After Lapsafter-laps: Practice sessions open with warmup laps and out-laps that aren't representative of race pace. Setting After Laps to 4 (the default) ignores each driver's first 4 laps before computing Best-N or Best-Consecutive. Raise it for sessions with long out-laps; lower for sessions where drivers got right to work.

Missing a term? Send feedback via the Contact form.