Win Rate Turn card was gained (only on your turn)

### Understanding the graph

#### Win rates

Win rate is a generalization of winning probability for multiplayer games.

A win in a 2 player game is worth 2 points, a 3 player game is worth 3, and so on. Someone who plays only 2 player games and wins every time will have a winrate of 2; a 2-player gamer who loses every game has a winrate of 0; a 2-player gamer who wins exactly as often as expected (half wins, half losses) has a winrate of 1.

The win rate of an event is the average number of points for that event. So an event that is uncorrelated with winning will have a win rate of 1.0, and events correlated with winning have a win rate of > 1.0, and losing with < 1.0.

#### Error bars

The error bars represent a 95% confidence interval in the measure for a particular event.

The smaller the bars, the more times a particular event was observed, and hence the less likely the true value is signficantly away from the center point due to randomness.

The error bars represent ±2 standard deviations. They encode the true mean for any particular point with a probability of about 95%.

#### Correlation

Don't get too carried away with interpreting these graphs: correlation does not imply causation.
• In general, cards with high costs tend to look good, because they are correlated with their buyer having lots of money. Someone who is able to get to 8 is naturally going to have a higher win rate than someone who is only able to get to 6.0
• Similarly, Duchies are often bought in the late game by players who are behind and unwilling to end the game by buying out the Provinces.
• In addition, the 'card advantage' graph is most accurate for non-terminal cards and cards that you typically want more than one copy of. For instance, it buckets the [2 Chapels vs 1 Chapel] situation into the same event as [1 Chapel vs 0 Chapels]. This is why, on average, the graph displays a 1-Chapel advantage has being correlated with a below-average winrate

#### The best choice

Note that these statistics are merely an average over many games. They do not necessarily say much about any particular game. So even though one card's early winning rate may be higher than a different card's winning rate on the same turn, it does not follow that the more winning card is neccesarily the best choice for any particular game.

#### Curses and Embargo

The apparent weirdness with the Curse graph is due to the fact that Curses are rarely intentionally gained. The only exceptions are when players are ending the game (almost always in a winning position) by running out the Curse pile, or are gaining Curses as a side effect of Embargo tokens. When people buy things despite being Embargoed, they are often buying a very valuable card.

### Tips on using the card filters

List different card names or card filters separated by a comma. Each filter is a JavaScript expression that is matched against every card. The expression is evaluated with attributes of the card filled as local variables. If this sounds confusing, don't worry. Here are some examples:

• Expansion=="Dominion" || Expansion=="Intrigue"

• Cost==7 returns Bank, Expand, Forge, and King's Court
• Cost == "P2" returns Apothecary, Scrying Pool, and University
• (""+Cost).indexOf("P")>=0 returns all cards with Potion in cost

• Actions >= 2 returns all cards that provide 2 or more Actions
• Cards < 2 returns all cards that draw either 0 or 1 Card when played
• Coins == "?" returns Bank, Philosopher's Stone, Pirate Ship, Salvager, Secret Chamber, Trade Route, Tribute, and Vault