Favorite vs Underdog Betting: Which Offers Better Value?

Neither side is automatically better value. Favorite vs underdog betting comes down to one comparison: the price a sportsbook offers against the true chance of that team winning. A favorite is the side the market considers more likely to win, and an underdog is the side it considers less likely, as covered in what the favorite and underdog labels mean. Favorites win more often but pay less per dollar staked; underdogs win less often but pay more. Whether either is worth backing depends on whether the payout is large enough for the risk, not on the label.

This article explains how a bettor can think about that comparison. It shows how each price converts into a break-even rate, why the sportsbook margin affects both sides, how a personal probability estimate changes the answer, and why underdog results swing so much more from bet to bet. The numbers are hypothetical examples, not current lines, and nothing here suggests that backing one side will produce a profit.

Price Versus Probability: The Real Question

The popular question “are underdogs better value?” assumes the label carries information about value. It does not. Value is a relationship between price and probability, not a property of a team. A heavy favorite can be good value if the price implies a lower chance than the real one, and a long underdog can be poor value if the payout fails to compensate for how rarely it wins.

Two terms matter here. The implied probability of a price is the win rate a bettor must achieve to break even at that price, and the fair probability is the bettor’s best estimate of the true chance. Value exists only when the fair probability is higher than the implied probability. This is the same idea described in expected value, applied to the choice between two sides of one event.

Because a sportsbook builds a margin into every market, both sides of a two-way market are priced slightly worse than fair. The two implied probabilities add to more than 100%. That means a bettor is not choosing between a “safe, fairly priced” side and a “risky, fairly priced” side; both sides start with a small built-in cost, and any edge must be large enough to overcome it.

Researchers and bettors have also discussed whether margin is spread evenly. The favorite-longshot bias describes the possibility that long odds carry a larger share of the margin than short odds. The evidence varies by market, so it is a reason to check the price rather than a rule that favorites are always the better buy.

How Prices Translate Into Break-Even Rates

Every American odds price converts to a break-even rate using the formulas for implied probability. For a favorite with negative odds, the rate is |odds| ÷ (|odds| + 100). For an underdog with positive odds, it is 100 ÷ (odds + 100). The break-even rate is the minimum win rate needed to avoid losing money over many equivalent bets, before any other considerations.

The table below applies those formulas to four hypothetical prices, using a $100 stake. Profit excludes the returned stake, so the payout in each case is the stake plus the profit shown.

Reading the table, each step toward a bigger payout raises the standard on the win rate in the opposite direction: a -300 favorite must win three times out of four to break even, while a +300 underdog needs only one time in four. Neither requirement is easier in any absolute sense. The question is always whether the real chance is above the number in the third column.

This is why comparing win rates alone misleads. Saying that favorites win more often is true and also irrelevant to value, because a higher win rate is already priced into the lower payout.

Worked Example: One Market, Two Opinions

Suppose a sportsbook lists a two-way market with the favorite at -200 and the underdog at +170. These are hypothetical prices, not a current market. The favorite’s break-even probability is 200 ÷ (200 + 100) = 66.67%, and the underdog’s is 100 ÷ (170 + 100) = 37.04%. Together they total 103.70%, so the market carries a margin of 3.70 percentage points.

Removing that margin proportionally gives fair probabilities of 66.67 ÷ 103.70 = 64.29% for the favorite and 37.04 ÷ 103.70 = 35.71% for the underdog. On a $100 stake, the favorite wins $50 (payout $150) and the underdog wins $170 (payout $270). At those fair probabilities, either side loses about $3.57 per $100 staked: 0.6429 × $50 − 0.3571 × $100 = −$3.57 for the favorite, and 0.3571 × $170 − 0.6429 × $100 = −$3.57 for the underdog. The margin is the price of playing, and it applies equally here.

Now add two opinions. Bettor A estimates the favorite wins 68% of the time. On the favorite, the expected value per $100 is 0.68 × $50 − 0.32 × $100 = +$2.00. On the underdog, with a 32% estimate, it is 0.32 × $170 − 0.68 × $100 = −$13.60. Bettor A’s estimate favors the favorite, if the estimate is right.

Bettor B estimates the favorite wins only 62% of the time. The favorite’s expected value becomes 0.62 × $50 − 0.38 × $100 = −$7.00, while the underdog’s, at 38%, becomes 0.38 × $170 − 0.62 × $100 = +$2.60. The same market produces opposite conclusions from different estimates. Neither bettor is guaranteed to be correct, and a two-point disagreement in probability is well within the normal error of any estimate.

How to Evaluate a Favorite or Underdog Price

A practical evaluation starts with the question “what is the win probability?” rather than “which side do I like?” The bettor first forms an estimate, ideally before looking at the price, using the approach described in estimating your own probability. Then the estimate is compared with the break-even rate. Only a gap in the bettor’s favor that exceeds the margin and the estimate’s own uncertainty is a reason to consider the bet at all.

It helps to think in ranges instead of single numbers. If Bettor B above believed the favorite’s true chance was somewhere between 60% and 66%, the underdog would look attractive only at the low end of that range and unattractive at the high end. A range that straddles the break-even rate is a signal that the estimate is not precise enough to justify either side, and passing is a legitimate decision.

It also helps to compare the same side across prices. A favorite at -150 and the same favorite at -170 are different bets with different break-even rates, 60.00% and 62.96%. Small price differences change the required win rate, which is one reason shopping for the best available price matters for both sides of a market.

Finally, a bettor’s edge has to survive the opposite-side check. If an estimate says both the favorite and the underdog are good value at once, the estimate is inconsistent, because the two probabilities must add to 100%. That is a sign to revisit the assumptions, not to bet both.

Common Mistakes When Comparing Favorites and Underdogs

The first mistake is treating the underdog payout as the reward for being bold. A larger payout does not mean greater value; it is the compensation for a lower chance, and a +300 price that wins only 20% of the time is worse than a -150 price that wins 65% of the time, despite the exciting payout.

The second is assuming favorites are the “safe” side. A favorite wins more often, but a loss costs more than a win earns. At -200, a bettor needs to win about two of every three bets just to break even, and a single loss erases two wins’ profit. Safer in frequency is not safer in expected result.

The third is generalizing from a streak of results. A run of underdog covers or favorite wins over a few weeks is a small sample, and small samples are dominated by chance. Past results should be tested against the price paid, not remembered as proof that one side is better.

The fourth is ignoring the margin. Many arguments for “always back dogs” or “always back chalk” quietly assume the prices are fair. They are not, so any blanket rule starts with a built-in loss before it considers how the teams actually match up.

The fifth is raising stakes to recover from losing streaks, which is especially tempting after a run of underdog losses. A past loss does not change the next outcome’s probability, and increasing stakes to recover losses only raises the risk of a larger drawdown.

Variance: How Each Side Feels Over Many Bets

Even when the estimated edge is identical, the two sides behave differently over time. Using the hypothetical numbers from earlier, suppose Bettor A backs the favorite at -200 with a true win rate of 68%, and Bettor B backs the underdog at +170 with a true win rate of 38%. The underdog bettor loses 62% of bets while the favorite bettor loses 32%.

That difference compounds. The chance of losing five bets in a row is 0.62⁵, about 9.2%, for the underdog bettor, and 0.32⁵, about 0.3%, for the favorite bettor. Long losing streaks are far more common on the underdog side, even if the long-run edge is real. This is variance at work: results spread widely around the expected value.

The spread can be measured. Per $100 bet, the favorite scenario has an expected profit of $2.00 and a standard deviation of about $70, while the underdog scenario has an expected profit of $2.60 and a standard deviation of about $131. Over 100 such bets, the expected profits are roughly $200 and $260, but the standard deviations are about $700 and $1,311. The underdog expectation is dwarfed by the noise, so a bettor could be down substantially after 100 bets with a sound estimate, or up substantially with an unsound one.

For responsible bankroll planning, this means a stake that feels comfortable on favorites may be too large on longer odds. Betting only money that can be lost, and setting limits before betting, matters more as the swings grow.

Where to Go Next

The practical takeaway is that the right question is “is this price better than my estimated chance?” and not “should I bet favorites or underdogs?” For readers building a framework, a natural sequence is to learn how a price becomes a probability, practice forming estimates, and then judge results over a large sample rather than a few weeks.

Passing on a market is also a result. Many markets will show no clear gap between an honest estimate and the price on either side, and not placing a bet in those cases is consistent with good decision-making. Nothing in this article suggests a bettor should wager more often; it is meant to help judge individual prices more clearly.