The harmonic convergence between my love of board games, coding, data analysis, and the holiday break has resulted in the following nerd-worthy dissertation on the Founding Fathers board game.
Founding Fathers is a board game designed by Christian Leonhard and Jason Matthews, and published by Jolly Roger Games.
This analysis is based on games played on Yucata.de, my favorite online board game site. All games played on Yucata.de are publicly available for replay and the format used is not overly difficult to parse for analyses like this one.
I used all games played by the top 10 rated Founding Fathers players (including #9, DangerDon :)), and all games played by the ten players with the highest number of Founding Fathers games played. This adds up to 922 unique games, 413 unique players, and 104,698 player moves.
The first question most players seem curious about is what makes a good score in Founding Fathers.
The chart is separated into 3-player, 4-player and 5-player games. Most of the games on Yucata.de are 3-player, followed by 4-player and 5-player. The total points scored by all players in a game is pretty much the same regardless of the number of players in the game. The more players in the game, the more players to divvy up the points, and the lower the scores. Winners in 3-player games average 32 points, winners in 4-player games average 26, and 5-player winners average about 22 points. The highest number of points in any of the games was 62.
The optimal number of influence markers for a player is another question that comes up. Everyone starts with three markers and can add more during the game. Here’s how it plays out in the actual games.
The fewer players in the game, the more turns you will play per round, so the more markers you will need. In addition to the number of players in a game, this chart is also divided into “Winner” and “Runner-up” categories. Winners are players that won the game or tied for first; runners-up are everyone else. On average, winners make do with slightly fewer influence markers than the runners-up.
At the end of a game of Founding Fathers, the factions associated with the 12 passed articles determine the bonus points awarded for debate tokens. Does the breakdown of the passed articles differ from the default (historic) breakdown? Here is the default breakdown:
Here is the actual breakdown in the played games:
And the answer to our question is… uh, no. After all of the player machinations, the passed articles break down pretty much the same as the historic articles.
That’s not to say that all articles get passed:
These “Convention” articles are the eight articles that are acted upon by the players: four articles in the Assembly Room and four articles in the Committee Room. There is a strong preference for players to pass an article in the Assembly room because influence markers on the Yea side get two points apiece if the delegate faction is the same as the article faction. However, there are reasons to vote Nay, including influencing the faction breakdown for debate bonus points, getting event points for voting Nay, and stopping an opponent from getting “Yea” points. The first four articles, the Virginia Plan, get flipped much more rarely:
Virginia Plan articles can only be flipped by William Samuel Johnson.
Each player in Founding Fathers is randomly assigned one of the five Planners. Do any of the Planners have an advantage over the others? Each of the Planners are from different states, and they have no faction affiliation. The player with James Madison always goes first, while the remaining order of the players is randomly assigned.
Going first is clearly an advantage in 3 and 5-player games, but not 4-player games. I was hoping to make some conclusions about the remaining Planners, thinking being associated with a particular state would be an advantage or disadvantage. However, the data didn’t cooperate. Sherman beats Patterson in 3-player games, but in 5-player games the situation is reversed. Pinckney has the best winning percentage in 4-player games.
The meat of a Founding Fathers game is the choices that players make with the delegate cards in their caucus (hand). On average, players perform 36 actions each in a 3-player game, 28 in a 4-player game, and 24 in a 5-player game. Note that this is not the same as the number of turns. For example, when a player chooses to play three delegates of the same faction at one time to move a debate marker up three spaces, I count that as three actions. The player has chosen to use a delegate for a debate action instead of a voting action or an event for three different delegate cards. With a given delegate card or cards, a player can execute an event, vote, debate, or snub (discard). Here is how the actions break down for runners-up and winners:
Both winners and runners-up play events for about 35% of their total actions. But, as we will see below, the choices of which events to play do differ between winners and runners-up.
Winners emphasize voting more than runners-up, 37% to 31%, and emphasize debating less. This matches my experience with the game. While you can’t ignore the end-of-game debate bonuses, a voting marker is a good investment. If you’re on the winning side, that marker can net you multiple points. Even if you’re on the losing side, you might get points from the committee room. You get no points for losing debate markers and you get no points for spending multiple turns pushing a debate marker up a track. There are events that reward you for voting Yea or Nay; there are no events that reward you for having debate markers.
You could argue that each time you push your marker up a debate track, you’re forcing an opponent to use up a turn to counter your move. Beware the situations where you and another player are battling over a debate track while everyone else is picking up points in the assembly room.
Note that the “Debate” actions in these two pie charts do not include events that affect the debate track. Those are Events.
Speaking of Events, the following table shows how players play each of the 55 different delegate cards. Players have four choices: Event, Vote, Debate, or Snub. The table shows the percentage of plays for each of those four choices. For a given delegate, those four numbers will add up to 100%. The delegates are sorted from highest Event percentage to lowest. The asterisk in Event Abbr denotes persistent events.
Delegate Name | State | Faction | Event Pct | Vote Pct | Debate Pct | Snub Pct | Event Abbr |
---|---|---|---|---|---|---|---|
John Langdon | NH | Fed | 72.9% | 18.3% | 8.5% | 0.3% | Caucus is 4* |
Pierce Butler | SC | Large | 67.4% | 23.1% | 8.4% | 1.1% | +2 Large Debate |
David Brearly | NJ | Small | 64.8% | 18.5% | 16.0% | 0.8% | Claim Unclaimed Debates* |
William Paterson | NJ | Planner | 64.4% | 10.0% | 25.0% | 0.6% | +1 Marker |
Roger Sherman | CT | Planner | 64.2% | 11.4% | 24.2% | 0.2% | +1 Marker |
James Madison | VA | Planner | 64.0% | 10.7% | 24.8% | 0.5% | +1 Marker |
Alexander Martin | NC | Antifed | 63.8% | 23.4% | 11.6% | 1.2% | +2 Antifed Debate |
John Dickinson | DE | Fed | 63.0% | 22.5% | 13.3% | 1.3% | +2 Fed Debate |
William Livingston | NJ | Small | 63.0% | 25.1% | 11.0% | 0.9% | +2 Small Debate |
Charles Pinckney | SC | Planner | 61.1% | 12.9% | 25.6% | 0.4% | +1 Marker |
Benjamin Franklin | PA | Large | 59.0% | 18.2% | 21.1% | 1.7% | +2 Points / Voted State |
Alexander Hamilton | NY | Planner | 56.3% | 14.3% | 28.9% | 0.4% | +1 Marker |
Gouverneur Morris | PA | Fed | 53.6% | 13.8% | 29.6% | 3.0% | Vote Plus 2nd Action |
Jacob Broom | DE | Fed | 53.0% | 17.3% | 28.0% | 1.7% | 2nd Vote |
Elbridge Gerry | MA | Antifed | 43.3% | 30.0% | 25.0% | 1.7% | +1 / Nay Vote* |
Thomas Mifflin | PA | Large | 42.8% | 28.6% | 27.7% | 0.9% | Play Draw Pool Event |
Daniel of St. Thomas Jenifer | MD | Fed | 41.9% | 22.4% | 33.0% | 2.7% | +1 if Yea |
James McClurg | VA | Fed | 39.0% | 30.5% | 29.0% | 1.4% | +1 / Fed Caucus |
George Washington | VA | Fed | 38.5% | 27.5% | 32.2% | 1.8% | End Round Now |
George Mason | VA | Antifed | 36.0% | 31.8% | 30.6% | 1.6% | +1 Marker for Committee Antifed |
Jonathan Dayton | NJ | Small | 34.8% | 38.5% | 26.1% | 0.6% | All Others Debate – 1 |
William Pierce | GA | Small | 33.1% | 36.6% | 29.2% | 1.1% | Use Top Discard |
George Wythe | VA | Fed | 32.7% | 30.7% | 35.8% | 0.8% | +1 Point / 3 Markers |
Luther Martin | MD | Antifed | 32.3% | 31.3% | 34.6% | 1.8% | No More Debates* |
John F. Mercer | MD | Antifed | 29.5% | 33.2% | 34.5% | 2.8% | +1 / Antifed Caucus |
Oliver Ellsworth | CT | Small | 28.9% | 41.2% | 28.8% | 1.2% | +1 / Small Caucus |
William Blount | NC | Antifed | 28.1% | 38.8% | 30.3% | 2.8% | +3 Points if Behind |
George Reed | DE | Fed | 27.6% | 33.2% | 38.2% | 1.0% | Remove 2 Voted Fed Delegates |
Richard Basset | DE | Small | 27.2% | 36.0% | 35.6% | 1.2% | Steal Delegate |
John Rutledge | SC | Fed | 26.9% | 38.3% | 33.8% | 1.0% | +1 Point / GA,NC,SC Winners* |
Rufus King | MA | Large | 25.8% | 41.4% | 30.3% | 2.5% | +2 Markers to Committee |
George Clymer | PA | Fed | 25.2% | 31.5% | 41.3% | 2.0% | +1 / Fed Considered Articles |
Robert Yates | NY | Antifed | 25.0% | 42.7% | 30.8% | 1.5% | Flip Assembly Article |
Nicholas Gilman | NH | Fed | 24.9% | 41.8% | 32.3% | 0.9% | +2 Points if Final Round |
Caleb Strong | MA | Fed | 23.8% | 38.8% | 36.3% | 1.2% | Use Available Delegate |
John Blair | VA | Fed | 21.8% | 41.2% | 35.8% | 1.2% | +1 for Voted Virginia |
Jared Ingersoll | PA | Antifed | 21.6% | 37.0% | 38.8% | 2.6% | Block all Events* |
James McHenry | MD | Fed | 21.6% | 35.1% | 40.5% | 2.8% | Replace MD |
John Lansing Jr. | NY | Antifed | 19.2% | 44.9% | 33.9% | 2.0% | +1 / Antifed Considered Articles |
William Samuel Johnson | CT | Small | 19.0% | 46.7% | 33.2% | 1.1% | Flip Resolved Article |
William Houston | NJ | Small | 18.6% | 42.8% | 37.4% | 1.3% | Discard Voted Delegate |
Abraham Baldwin | GA | Small | 17.2% | 43.0% | 37.7% | 2.1% | +1 For Voted CT |
Daniel Carroll | MD | Large | 15.9% | 41.4% | 39.5% | 3.2% | Override MD |
Hugh Williamson | NC | Small | 15.2% | 43.9% | 39.2% | 1.7% | +1 to Committee / NC Delegate |
Gunning Bedford Jr. | DE | Small | 13.4% | 44.1% | 39.9% | 2.6% | Remove 2 Small Voted Delegates |
Edmond Randolph | VA | Fed | 12.8% | 43.3% | 42.6% | 1.2% | Flip Voted State |
Richard Dobbs Spaight | NC | Fed | 11.6% | 43.1% | 43.3% | 2.0% | Your Markers Protected* |
William Richardson Davie | NC | Small | 10.5% | 48.9% | 39.7% | 1.0% | Can’t Override Your Delegates* |
Charles Coteworth Pinckney | SC | Fed | 9.4% | 49.3% | 39.6% | 1.7% | -1 Point / voted GA,NC,SC |
Thomas Fitzsimons | PA | Fed | 8.3% | 37.8% | 50.9% | 3.0% | Discard Voted Delegates to 1 |
William Houstoun | GA | Large | 8.0% | 54.0% | 35.9% | 2.0% | Discard Voted GA |
James Wilson | PA | Large | 6.6% | 45.0% | 44.7% | 3.6% | Wilson Votes Either Side |
Robert Morris | PA | Fed | 6.1% | 39.6% | 51.1% | 3.2% | No Washington for Others* |
William Few | GA | Fed | 6.0% | 49.8% | 42.2% | 2.0% | Discard Draw Pool |
Nathaniel Gorham | MA | Large | 3.5% | 54.2% | 39.3% | 2.9% | Remove Other Committee Markers |
The delegates at the top of the table are those that get played as Events the most. In the top 12 delegates, we see all five planners, which are used to gain influence markers. Madison, Paterson, and Sherman are ranked above Pinckney and Hamilton because the last two planners can appear in games unattached to a player, making it more likely that they won’t be needed for additional markers.
In the top 9, we see all four events that move a debate marker up two spaces.
At the #11 spot is Benjamin Franklin, an event that can gain a lot of points at once.
On top of the list is John Langdon, a persistent event that allows players to have a caucus size of 4 until the end of the round.
At the bottom of the list is Nathaniel Gorham, an event that removes other players’ committee markers. Why is this event on the bottom? Taking away the two event cards that add markers to committee, to be in a position to use Gorham, you would have had to have markers on the losing side in Assembly, and have not scored for those markers because you didn’t have the most makers in committee, and have other player(s) in the same situation. Being in a position to use Gorham is not, shall we say, aspirational. What information can we gather about the various events in this table? How players perceive the value of an event certainly affects the event’s position, but there are at least two other factors at work: legality and timing. The top 12 events can all be legally played at any time (except when Jared Ingersoll is out, which blocks all events). Most of the top 12 events are also useful at any time, with Benjamin Franklin requiring the most setup. That still doesn’t tell us as much as we’d like about which events lead to more won games.
The next table compares how winners play the events compared with runners-up.
Delegate Name | State | Faction | Event Pct by Winners | Event Pct by Runners-up | Delta (Winner Preference) | Event Abbr |
---|---|---|---|---|---|---|
Jacob Broom | DE | Fed | 64.1% | 46.6% | 17.5% | 2nd Vote |
Daniel of St. Thomas Jenifer | MD | Fed | 53.4% | 36.7% | 16.6% | +1 if Yea |
Benjamin Franklin | PA | Large | 67.9% | 53.6% | 14.3% | +2 Points / Voted State |
Gouverneur Morris | PA | Fed | 63.0% | 49.2% | 13.8% | Vote Plus 2nd Action |
George Washington | VA | Fed | 46.7% | 34.7% | 12.0% | End Round Now |
George Mason | VA | Antifed | 43.8% | 32.4% | 11.4% | +1 Marker for Committee Antifed |
David Brearly | NJ | Small | 70.9% | 61.6% | 9.3% | Claim Unclaimed Debates* |
John F. Mercer | MD | Antifed | 35.2% | 26.9% | 8.3% | +1 / Antifed Caucus |
James McClurg | VA | Fed | 44.6% | 36.4% | 8.2% | +1 / Fed Caucus |
Daniel Carroll | MD | Large | 20.5% | 13.8% | 6.7% | Override MD |
John Rutledge | SC | Fed | 31.4% | 24.8% | 6.6% | +1 Point / GA,NC,SC Winners* |
Nicholas Gilman | NH | Fed | 28.7% | 23.2% | 5.5% | +2 Points if Final Round |
William Pierce | GA | Small | 36.8% | 31.5% | 5.3% | Use Top Discard |
James McHenry | MD | Fed | 25.0% | 20.2% | 4.8% | Replace MD |
Oliver Ellsworth | CT | Small | 31.5% | 27.7% | 3.8% | +1 / Small Caucus |
Caleb Strong | MA | Fed | 25.9% | 22.8% | 3.1% | Use Available Delegate |
George Wythe | VA | Fed | 34.0% | 32.1% | 1.8% | +1 Point / 3 Markers |
George Reed | DE | Fed | 28.8% | 27.0% | 1.8% | Remove 2 Voted Fed Delegates |
George Clymer | PA | Fed | 26.1% | 24.8% | 1.4% | +1 / Fed Considered Articles |
James Wilson | PA | Large | 7.3% | 6.2% | 1.1% | Wilson Votes Either Side |
Elbridge Gerry | MA | Antifed | 43.7% | 43.1% | 0.6% | +1 / Nay Vote* |
Abraham Baldwin | GA | Small | 17.3% | 17.2% | 0.2% | +1 For Voted CT |
Jared Ingersoll | PA | Antifed | 21.5% | 21.7% | -0.2% | Block all Events* |
Robert Morris | PA | Fed | 5.8% | 6.2% | -0.4% | No Washington for Others* |
Thomas Mifflin | PA | Large | 42.3% | 43.0% | -0.7% | Play Draw Pool Event |
Gunning Bedford Jr. | DE | Small | 12.9% | 13.7% | -0.7% | Remove 2 Small Voted Delegates |
William Richardson Davie | NC | Small | 9.9% | 10.7% | -0.8% | Can’t Override Your Delegates* |
Edmond Randolph | VA | Fed | 12.0% | 13.2% | -1.2% | Flip Voted State |
John Langdon | NH | Fed | 71.6% | 73.5% | -1.9% | Caucus is 4* |
John Lansing Jr. | NY | Antifed | 17.8% | 19.9% | -2.0% | +1 / Antifed Considered Articles |
Thomas Fitzsimons | PA | Fed | 6.5% | 9.1% | -2.6% | Discard Voted Delegates to 1 |
William Houstoun | GA | Large | 6.1% | 8.8% | -2.7% | Discard Voted GA |
Nathaniel Gorham | MA | Large | 1.6% | 4.3% | -2.7% | Remove Other Committee Markers |
Richard Dobbs Spaight | NC | Fed | 9.8% | 12.5% | -2.7% | Your Markers Protected* |
William Few | GA | Fed | 3.6% | 6.9% | -3.4% | Discard Draw Pool |
Richard Basset | DE | Small | 24.7% | 28.3% | -3.6% | Steal Delegate |
Robert Yates | NY | Antifed | 22.4% | 26.2% | -3.8% | Flip Assembly Article |
Luther Martin | MD | Antifed | 29.3% | 33.5% | -4.3% | No More Debates* |
John Blair | VA | Fed | 18.5% | 23.1% | -4.6% | +1 for Voted Virginia |
Roger Sherman | CT | Planner | 60.7% | 65.5% | -4.8% | +1 Marker |
Pierce Butler | SC | Large | 64.1% | 69.1% | -5.0% | +2 Large Debate |
William Samuel Johnson | CT | Small | 15.2% | 20.5% | -5.3% | Flip Resolved Article |
Jonathan Dayton | NJ | Small | 30.9% | 36.5% | -5.7% | All Others Debate – 1 |
Hugh Williamson | NC | Small | 10.9% | 16.7% | -5.8% | +1 to Committee / NC Delegate |
Rufus King | MA | Large | 21.8% | 27.7% | -5.9% | +2 Markers to Committee |
Alexander Martin | NC | Antifed | 59.2% | 65.7% | -6.5% | +2 Antifed Debate |
Charles Coteworth Pinckney | SC | Fed | 4.9% | 11.4% | -6.5% | -1 Point / voted GA,NC,SC |
William Houston | NJ | Small | 13.7% | 20.8% | -7.1% | Discard Voted Delegate |
James Madison | VA | Planner | 59.1% | 66.2% | -7.1% | +1 Marker |
Alexander Hamilton | NY | Planner | 51.1% | 58.5% | -7.4% | +1 Marker |
William Livingston | NJ | Small | 57.8% | 65.6% | -7.8% | +2 Small Debate |
William Paterson | NJ | Planner | 58.5% | 67.0% | -8.5% | +1 Marker |
John Dickinson | DE | Fed | 57.0% | 65.6% | -8.6% | +2 Fed Debate |
Charles Pinckney | SC | Planner | 55.0% | 63.9% | -8.9% | +1 Marker |
William Blount | NC | Antifed | 21.7% | 31.1% | -9.5% | +3 Points if Behind |
Event Pct by Winners shows the percentage of time that a winning player (one that eventually won the game) chose to play a delegate’s event instead of the other three choices. The Event Pct by Runners-up column shows the same data for the Runners-up. Delta (Winner Preference) is the percentage point difference between these two numbers and indicates events that winners prefer more than other players (Delta is positive) and those events that winners prefer less than other players (Delta is negative). The delegates are sorted by highest delta to lowest.
Delegates at the top of the list are not necessarily the “best” cards; they’re the events that winners play more often than runners-up and so deserve some consideration. An approach to improving one’s games is to try to emulate the event percentages played by the winners, where appropriate.
The delegates ranked #1, #2, and #5 are notable because it’s easy to misplay these events on Yucata.de, andeach of these events are almost always useful to execute. Morris allows you to vote than take a 2nd action, Broom allows you to vote then vote a 2nd time, and Daniel allows you to vote and get an immediate point if the vote is Yea. To get the events to work, you have to execute the event, then vote. These aren’t triggered events like in Magic; you have to specifically choose the event.
My assumption is that the appearance of these delegates near the top of the Winner Preference table is because runners-up are misplaying these events more than the winners and are missing out on scoring opportunities. I’ve misplayed a couple of those events myself when I first started playing on Yucata.de.
Benjamin Franklin, #3 on the list, scores two points for each of the player’s markers in the assembly room, then removes the markers. Franklin is a potentially high-scoring card, but it takes patience and timing to get the most out of it. You want to wait until you have as many markers in play as possible without having the round end before you’ve had a chance to play Franklin. Winners are able to play Franklin’s event at a rate of 68% compared to 54% for runners-up.
Four of the seven delegates at the bottom of this list are Planners. We’ve already seen in the Influence Markers chart above that winners make do with fewer influence markers on average than runners-up.
Now it’s time to talk about the Father of Our Country, #5 on the list. Washington ends the current round immediately, and the assembly article passes or fails based on the votes already cast. Washington is the most disruptive event in the game and causes the most game-balance debates. Winners use Washington at a 47% rate compared to a 35% rate for runners up.
I always look for an opportunity to get Washington as much to keep it out of others hands as to use myself. I’ve been able to use Washington a number of times at the end of games, and when it works, it is indeed glorious.
The Washington debate is usually about whether or not the card is unbalanced or broken. My conclusion: absolutely not, for a couple of reasons. First, to use Washington to end the game in your favor, you have to be ahead in projected final points after everyone else has just made a move. If winners are able to use Washington at a 47% rate compared to 35%, it could be because they tend to be leading in games.
The second factor that makes Washington (and all other events, for that matter) less powerful is semi-secrecy: the states and factions of everyone’s hands are known and full information is available for all discard cards. As a player, you know when no one else has Washington (he’s in the discard pile or assembly room, or no one has a Virginia Fed), you know when someone definitely has Washington (someone has a Virginia Fed and all other Virginia Feds are accounted for), and you know when someone might have Washington (someone has a Virginia Fed).
Respectfully submitted,
DangerDon (Don Laabs)
P.S. If you’d like to comment on the analysis, you can do so on this BoardGameGeek thread.
You must be logged in to post a comment.