🏈 The 1870 college football season

The champion and its path, the poll race week by week, the honours, the best games and every conference table.

Source: Citizen of Nowhere college football database · 1 conferences · 0 bowls
Jump to any season1869–2026
1870season

1870 national champion

Named by the selectors, with the road it took.

National champion
1-0NCF

The path, Princeton1-0

  1. Nov 12vs
    Rutgers
    W6-2

The 1870 Elo race

Elo by week, higher is better. There was no AP poll this year, so the ratings tell the story.

How this is measured
The chart plots the 15 highest-rated teams at the end of the season, sampled each Sunday from the first game to the last. Ratings are on the site's college football Elo scale, where the average major team finishes the season at 1500. The final top five and the champion are drawn in team colours; tap a line or pick a team to follow one.
PrincetonRutgersColumbia3 teams · tap or drag for that week · tap a line to lock it

Tap or drag on the chart to read a team's Elo in any week.

1870 Elo at season end

There is no AP poll before 1936, so the ratings stand in: the ten highest-rated major teams when the season ended.

1870 Elo at season endElo

Elo top 3

rating, z
1Princeton1647 · z +0.9
2Rutgers1575 · z +0.5
3Columbia1278 · z −1.4

The best games of 1870

The ten best games of the season by Game Score. Switch to the Elo score for a second opinion.

How this is measured
Game score is the college football workbook's own rating of how good a game was; it weighs a lot more than the margin. The Elo score is a second view built from the ratings alone: the two teams' average pre-game Elo, plus stakes (150 for the national championship game, 90 for a playoff game or a major bowl, 40 for another bowl or a conference title game, 15 for a rivalry), plus up to 60 for a close finish and 15 for overtime. The Elo view takes its own top ten from every game between major teams. Sort by any column.
The best games of 18701 games
Rank by
1
6-2 · 1870-11-12 · Elo score 1660
0.61
game score

Upsets of 1870

The results the ratings gave the winner the least chance of, lowest first.

How this is measured
Win chance is the winner's pre-game probability from the site's college football Elo ratings, home field included, so 3% means the winner would be expected to win about one time in thirty. Games against lower-level opponents count and carry a tag for the level (FCS, Division II and so on). Where the AP poll had ranks, the rank line is shown as context. Elo gap is the loser's pre-game rating minus the winner's.
Upsets of 18702 upsets
1
6-2 · 1870-11-12
63%
win chance
2
6-3 · 1870-11-05
89%
win chance

1870 standings by conference

Every conference at once: conference and overall record, and the final AP rank. A star marks the conference champion.

1870 standings by conference1 groups

Independent

3 teams
#TeamConfAll
1Princeton-1-0
2Rutgers-1-1
3Columbia-0-1

Where these numbers come from

Enough to disbelieve this page on purpose rather than by accident.

Champions, polls, standings, bowls, honours and games are read from the Citizen of Nowhere college football database, built from its own workbook one season at a time. Team colours and crests are the current programs’, so a school is drawn the way it looks today even where it played under another name.

The AP poll begins in 1936, so earlier seasons have no poll race and no final poll, but they do have an Elo race, and their champions are those the selectors named. Before 1992 there was no title game, so those seasons list the major bowls instead; from 1992 the title game is shown with the other major bowls, and from 2014 the playoff bracket. Where the database has no entry for a section, the section is left out rather than shown empty.

Elo ratings, the upsets and the best games come from the site’s college football Elo, which rates every game since 1869 and counts games against lower-level opponents. Each season’s ratings are shown on a scale where the average major team ends the season at 1500; z is the distance from that average in standard deviations.