When your tenth pick locks in, 17-0 turns your roster into one number, converts that number into a win probability, and plays out 17 games and a playoff run. Nothing about that process is hidden or mysterious, so this page lays it all out. Every figure below comes from the game's season engine as it stands in October 2026. If you only want the rules of the game, see How to Play. If you want drafting advice, see How to Build the Perfect NFL Roster. This page is about the machinery.
1. Player ratings
Every player card in the database carries one whole-number rating for one franchise and one era (a decade, from the 1960s to the 2020s). Ratings in the database run from 75 to 99. A rating is attached to that one franchise-and-era entry and reflects how dominant the player was at his peak, so it is not a career score. In Classic mode the rating is printed on the card. In Pro Scout mode it is hidden, along with career stats and accolades, and the spin's players are listed alphabetically so the order does not give the ratings away. The engine uses the same ratings either way.
2. Positional weighting
The engine does not take a plain average of your ten ratings. Each position has a weight, and your weighted average rating is the sum of (rating × weight) divided by the sum of the weights.
| Position | Weight | Share of the average |
|---|---|---|
| QB | 2.5 | 19.2% |
| EDGE | 1.5 | 11.5% |
| OL | 1.4 | 10.8% |
| WR | 1.3 | 10.0% |
| DT | 1.2 | 9.2% |
| CB | 1.2 | 9.2% |
| TE | 1.1 | 8.5% |
| RB | 1.0 | 7.7% |
| LB | 0.9 | 6.9% |
| S | 0.9 | 6.9% |
The weights add up to exactly 13.0, which makes the effect easy to read. One rating point at quarterback moves your average by 0.19. One rating point at safety moves it by 0.07. A quarterback is worth a little under three safeties, but note what the table does not say: the weights only shape the average, and the weak-link penalty in section 4 ignores position entirely.
3. The chemistry bonus
Your quarterback earns a bonus for every pass-catcher who shares his franchise and his era. A matching wide receiver adds +2 and a matching tight end adds +2. The code caps the total at +6, but a roster has only one WR slot and one TE slot, so the bonus you can actually reach is +4. The +6 cap is never triggered.
Two details are easy to miss. First, the match is exact: the franchise code and the decade must both be identical, so a 1980s San Francisco quarterback pairs with a 1980s San Francisco receiver, not a 1990s one and not a 1980s Raiders one. Second, the draft screen flags eligible WR and TE cards with a small “+2” chemistry badge, and that badge shows in Pro Scout mode too. It is the one rating-related hint Pro Scout does not hide.
What +4 is worth depends on where you are on the curve. For a roster that would otherwise sit at a team strength of 73.3, it lifts the strength to 77.3 and the per-game win probability from 70.1% to 72.6%. It is real but modest, which is why chemistry should break ties between similar players and not drive a pick on its own.
4. The weak-link penalty
The engine finds your single lowest-rated player, whatever his position, and subtracts a penalty from team strength. Only the lowest player counts, so two weak players cost no more than one.
87 ≤ lowest < 90: penalty = (90 − lowest) × 0.4
lowest < 87: penalty = (87 − lowest) × 1.5 + 1.2
| Lowest rating on roster | Penalty (strength points) |
|---|---|
| 90 or higher | 0 |
| 89 | 0.4 |
| 88 | 0.8 |
| 87 | 1.2 |
| 85 | 4.2 |
| 82 | 8.7 |
| 80 | 11.7 |
| 77 | 16.2 |
| 75 | 19.2 |
The penalty is the harshest force in the engine. Below 87, every point at your weakest spot costs 1.5 strength points, while a point at quarterback, the heaviest position, adds only 0.19 to the weighted average. The curve is continuous at 87 and 90, so there are no cliffs, just a steepening slope.
5. From team strength to win probability
per-game win probability = 0.40 + 0.48 × (strength / 100)1.5
| Team strength | Win probability per game | Expected regular-season wins |
|---|---|---|
| 70 | 68.1% | 11.6 |
| 80 | 74.3% | 12.6 |
| 90 | 81.0% | 13.8 |
| 95 | 84.4% | 14.4 |
| 100 | 88.0% | 15.0 |
Strength is capped at 100, so the best possible per-game win probability is 88%. No roster, however stacked, is ever a lock to win a given week. Every game is a separate weighted coin flip at the same probability; there are no opponent ratings, home-field effects or injuries in the model. The opponent abbreviation and the score shown in the game log are decoration generated afterward and have no effect on who wins.
6. How the season is settled, and why it repeats
The 17 regular-season results come from a seeded pseudo-random number generator. The seed is a hash of your roster: for each of the ten position slots, the player’s name, position, team, era and rating. Draft order is not part of the hash, because players are keyed by slot. That produces three properties you can rely on.
- Same roster, same season. Build the identical ten players in any order, in any mode, on any day, and you get the identical record and playoff run.
- Change one player, reshuffle everything. Swapping a single player changes the seed, so every game’s random draw changes, not just the one involving his position.
- Better does not guarantee luckier. Because each roster has its own fixed stream of draws, a slightly stronger roster can finish with a worse record than a slightly weaker one. Win probability is the right way to compare rosters before you simulate, not a promise about any single result.
The game also uses a date-based seed in one other place: the Daily Eye Test draws its franchise-and-era spins from the calendar date, so everyone gets the same ten spins that day. That is separate from the season simulation.
7. Playoffs, byes and what counts as 17-0
These thresholds are rules of this game, not NFL rules. After the 17 games:
- Fewer than 9 wins: no playoffs, and the season ends there.
- 9 to 13 wins: you enter at the Wild Card round and need four playoff wins: Wild Card, Divisional, Conference Championship, and the Championship.
- 14 or more wins: you earn a first-round bye and need three playoff wins.
Playoff rounds are single games, and one loss ends your run. Each later round is a little harder: the win probability for the round that is r steps into your path is your regular-season probability minus 0.02 × r, with a floor of 30%. At the 88% ceiling, a bye team faces 88%, then 86%, then 84%. The engine does not model home field or neutral sites, and a bye is simply a shorter bracket.
The game awards 17-0 only when both conditions are met: you won all 17 regular-season games and you won the championship. A 17-0 regular season that ends in a playoff loss is shown as a Super Bowl loss or an earlier exit, not a perfect season. That is a stricter test than the NFL record book applies. The 1972 Miami Dolphins are the only NFL team to win a title while going unbeaten, finishing 14-0 in the regular season and then winning three playoff games, which is why “17-0” fits them only when playoffs are counted. The 2007 New England Patriots won all 16 regular-season games and lost Super Bowl XLII, a story told in The 2007 Patriots.
8. Worked examples
The following are illustrative rosters, built so you can follow every step. The ratings for the first three players are real entries in the database; the rest are chosen to make a point.
Example A: a star-heavy roster with a hole
QB 99 (Joe Montana, SF 1980s), WR 99 (Jerry Rice, SF 1980s), TE 82 (Russ Francis, SF 1980s), RB 88, OL 84, EDGE 90, DT 77, LB 90, CB 79, S 98.
- Weighted sum: 99×2.5 + 99×1.3 + 82×1.1 + 88×1.0 + 84×1.4 + 90×1.5 + 77×1.2 + 90×0.9 + 79×1.2 + 98×0.9 = 1163.4. Divide by 13.0 and the weighted average is 89.49.
- Chemistry: Rice and Francis both match Montana, so +4.
- Weak link: the lowest rating is 77, so the penalty is (87 − 77) × 1.5 + 1.2 = 16.2.
- Team strength: 89.49 + 4 − 16.2 = 77.29. Per-game win probability: 0.40 + 0.48 × 0.77291.5 = 72.6%, or about 12.3 expected wins.
- Chance of 17-0 in the abstract: 17 straight wins is 0.72617, about 0.43%. Then three bye-round playoff wins at 72.6%, 70.6% and 68.6% multiply that by about 0.35, for roughly 0.15%, or 1 in 650.
Despite three legends at the top, the 77 at defensive tackle costs 16 points of strength, more than every chemistry and positional advantage combined.
Example B: the same stars, no hole
Keep the same quarterback, receiver, EDGE, LB and S, but raise TE, OL, DT and CB to 88 (RB stays at 88). The weighted average rises to 92.28, chemistry stays at +4, and the lowest rating is now 88, so the penalty is just 0.8. Team strength is 95.48 and the per-game win probability is 84.8% (about 14.4 expected wins). The odds of 17-0 improve to roughly 3.4%, or 1 in 29. Most of that jump came from removing the penalty (16.2 down to 0.8), not from adding rating points.
Reference points for the 17-0 odds
These odds are not measured from play. They are computed from the formulas above as 17 straight wins, times the three bye-round playoff wins, for a roster whose ten players all share one rating (so there is no weak-link gap). Because each real roster is fixed by its seed, read them as the chance that a roster of that quality produces a perfect season, not as a promise.
| Every player rated | Per-game win probability | Chance of 17-0 |
|---|---|---|
| 88 | 79.1% | about 1 in 120 |
| 90 | 81.0% | about 1 in 73 |
| 95 | 84.4% | about 1 in 32 |
| Strength 100 (the ceiling) | 88.0% | about 1 in 14 |
The 1-in-14 ceiling, about 7%, is the single figure to remember for the best case: perfect ratings, full chemistry and no weak link. Anything short of that is harder, and a roster with a hole like Example A is far harder. That is why a 17-0 is meant to feel rare.