Goal or Assist Bets Explained: Rules, Maths and When They're Worth It

You need your player to be credited with at least one goal or at least one assist in the match. One of either is enough. Doing both doesn't pay extra, since it's one bet with one outcome.

The two halves follow the normal rules for each market:

• The goal part works like an anytime goalscorer bet. Own goals don't count. Penalties scored in normal play do. Shootouts don't.
• The assist part works like a player to assist bet. Usually that means the standard data provider definition: the final touch from a teammate that leads to a goal. Winning a penalty, forcing an own goal or having a shot saved into a teammate's path normally don't count, even though they would earn a Fantasy assist.

These vary, so check your bookmaker's rules before you bet:

• Non-starters. Bet365's rules say player markets are void if the player doesn't start. Others may only void if he takes no part.
• Extra time. Bet365 says its player match markets include extra time if played. Others settle on 90 minutes plus stoppage time.
• Assist definition. Many settle on an official data provider's standard definition. If your bookmaker uses a broader definition, the market is easier to win and should be priced shorter.
• Disallowed goals. A goal ruled out by VAR is no goal and no assist.

This is where most people go wrong. If a player has a 33% chance to score and a 20% chance to assist, it's tempting to think goal or assist is 53%. It isn't, because some of the matches where he scores are also matches where he assists. Adding them counts those matches twice.

A cleaner way to think about it: goals and assists are both "goal involvements". If you expect a certain number of goals and a certain number of assists from him, add those expectations together, then work out the chance of at least one involvement.

Example (hypothetical). A forward has expected goals of 0.40 and expected assists of 0.22 for this match, with his expected minutes already built in.

The naive sum would make you think a price of 2.00 is generous. With the better estimate, 2.00 is actually below fair.

This model assumes goals and assists arrive independently, which isn't quite true. When his team has a big game, his goal and assist chances rise together. Real models handle that, but the main lesson holds: the combined chance is always less than the two added up.

Once you have a fair odd, the check is the usual one. Using the hypothetical player above, with a fair odd of 2.16:

• Bookmaker offers 2.00: EV = 0.462 x 2.00 - 1, about -7.6%.
• Bookmaker offers 2.30: EV = 0.462 x 2.30 - 1, about +6.3%.

A second useful check is consistency. If a bookmaker's goal price and assist price for the same player imply a combined chance well away from its goal or assist price, one of the three is probably off. That doesn't tell you which one, but it tells you where to look. The implied probability guide covers turning prices into chances.

Players whose involvement is spread across both columns: second strikers, attacking midfielders, wide forwards who both shoot and cross, and set-piece takers who also get into the box. For a pure finisher, most of the value of goal or assist is already in his goalscorer price. For a pure creator, it's mostly in his assist price.

You can't build "goal or assist" by combining a goal leg and an assist leg in a bet builder, because a builder needs every leg to win. Goal AND assist is a much less likely outcome, priced much bigger. If you want either one, use the goal or assist market. If you're adding it to a builder with other legs from the same match, read bet builder tips on correlation first.