When Both Sides Drop: What 126 Matches With Signals on Both Teams Show

What a "both sides" match is, and why it happens

An OddsDrop signal is triggered when the sharpest bookmaker cuts a price sharply: the market now rates that outcome as more likely than it did a few minutes ago. If Pinnacle shortens the home team from 2.60 to 2.10, that is a signal on the home team. Sometimes, an hour or two later, the market moves again, and the away team shortens by a similar amount. Now the same match has signals on both teams.

It looks contradictory, but there is nothing broken about it. A few common reasons:

• New information arrives in pieces. An early move on a rumoured line-up can be reversed when the confirmed team sheet shows the opposite.
• Informed participants disagree. Different syndicates and models can take different sides at different times. A sharp market absorbs both.
• Overreaction and correction. A large stake can push a price too far, and the correction can look like a drop on the other side.
• Thin markets. In lower leagues a small amount of money moves a price noticeably, and the price can swing both ways before kick-off.

The key point: our signal describes a market move, not a forecast of the result. Reporting both moves is the honest behaviour. Hiding the second one would make the record look cleaner, and it would no longer be a complete record.

How often it happens

We analysed every VIP football moneyline signal (full-time result, home or away side) logged up to 22 October 2026, grouped by match. Four matches whose results had been recorded impossibly (both teams marked as winners) were left out of every number below.

Inside those 126 matches, the signals are not always one against one:

So a pair of opposing signals is rare (about one match in thirty), but it is a normal part of how a live market behaves, and more than one in five of those cases has a repeat on at least one side.

Which signal came true more often?

The natural question is whether the first signal or the later one is more trustworthy. Of the 123 matches where every signal has been settled (we take the first signal and the first signal on the opposite side after it):

The later signal is ahead by three matches, which is not a result. With 123 observations the standard error of a win rate is about 4.5 percentage points, so 41.5% and 43.9% cannot be told apart. Splitting the data further points the same way:

• By time between the two signals. Within 30 minutes to 2 hours (40 pairs) it is an exact tie, 17 to 17. Between 2 and 6 hours (71 pairs) the later signal leads 33 to 28, again within noise. The other groups have fewer than ten pairs and say nothing.
• By size of the drop. The signal with the bigger drop won 36.2% of the time when it came first and 35.2% when it came second: no difference.
• By favourite and underdog. The side with the lower price won 48.8% of the time, the longer-priced side 36.6%. That simply restates that favourites win more often than underdogs.

The one number that stands out is the draw. In 14.6% of these matches neither team won, which means both signals lost. A 3-way market cannot be fully hedged by backing the two teams: the draw is the third outcome.

Does it pay? The result in units

This is the part most people care about. We priced every bet at the Suggested price published with the signal (1 unit per bet), the same basis as our public statistics, and compared four ways of betting the same 123 matches:

For comparison, the 3,693 settled matches signalled on one side only returned a 54.2% win rate and +11.2% ROI on the same basis.

How should that be read?

• Every row is statistically zero. At around 2.0 odds the standard error of ROI is roughly ±9 points on 123 bets and ±6 on 246. A result of −1.7% sits well inside that.
• But it is clearly below the one-sided signals. The gap to +11.2% is about two standard errors. Matches where the market moves both ways are, on this evidence, a place where the edge is weak or absent.
• Backing both sides is not a profit engine. It is easy to assume that holding both teams must lock in something. It does not, because the draw loses both bets (14.6% of these matches) and because both bets are placed at prices that already include the bookmaker's margin.
• The cost is small. These matches are 6.5% of all signals. Even a few units of drag over four months is negligible against a record of thousands of signals.

The sensible reading is neutral: they are not worth fearing and not worth avoiding, but they are not where you should add stake.

What about cashing out the first bet?

A common idea is to take the first bet and cash it out in profit once the opposite signal appears. Our data does not support that as a general rule, with an important caveat: only 25 of these pairs have a recorded closing price.

One-sided signals and the later signal of a pair behave alike: about a third of them close at a shorter price than the one we signalled. The earlier signal of a pair does worse, only one in five. Once the market has moved against the first team, that team's price has usually drifted back out, not further in. In plain terms, the position you took first is typically worth less at that moment, so a cash-out from price movement alone is more likely to be at a loss than at a profit. Your own bookmaker's price, margin and cash-out terms change the picture, so check it against your own records rather than assuming.

How to handle it in practice

OddsDrop is a market-monitoring tool, not an advisory service, and none of this is personal financial advice. These are the points the data supports:

1. Expect them. About one match in thirty gets both sides, and in more than one such case in five there is a repeat on at least one side. A contradictory pair is normal behaviour, not an error.
2. Do not increase the stake because there are two signals. Two opposing moves are two partial pieces of information that cancel each other, not double confirmation.
3. Do not assume that backing both sides locks in profit. A draw loses both bets, and each bet is priced with margin. Only a genuine arbitrage across all three outcomes guarantees a result, and signals do not guarantee that.
4. Do not rely on a cash-out to rescue the first bet. It depends on your bookmaker's cash-out price, which usually includes extra margin.
5. Keep your own records. Log which matches had signals on both sides and how you bet them. After a few hundred bets you will know more than any four-month sample of ours.
6. Stay with flat stakes. Because the expected result here is close to zero, variance matters more than edge. A smaller stake on contested matches is a reasonable personal choice, though our data does not prove that it is better.

What we do about it

• We publish every signal, including both sides of a match. They are in the public record and in the statistics. None is hidden or deleted.
• We measure them separately. This article is that measurement. Doing it also exposed four results that had been recorded impossibly (both teams marked as winners). We excluded them here and are checking how they got through. We would rather find our own errors than leave them in.
• We keep testing. Whether to hold back a second, opposing signal is a legitimate question. Today the evidence does not justify a change: the numbers are too small and too close to zero. We will repeat the analysis as the sample grows and publish what changes.

Conclusion

Signals on both teams of the same match are a fact of how markets move. They affect 3.3% of matches and 6.5% of signals. The first signal won 41.5% of the time, the second 43.9%, and 14.6% ended in a draw that beat both. At suggested prices the whole group is close to break-even (−1.7%), well below the +11.2% of one-sided signals, but too small a sample to prove a lasting difference. They cannot be avoided, they should not alarm you, and they are not a reason to change your stake. They are one more thing worth knowing when you follow dropping odds.