The EMA crossover strategy won 69% of its trades and still lost $9,035
Live paper record 14 Sep 2026 to 13 Oct 2026. Updated 13 Oct 2026.
Short answer: it won most of its trades and still lost a lot of money. Three versions of a 15-minute EMA crossover strategy took 39 paper trades on same-day SPX options between 14 Sep and 13 Oct. 27 of them made money, a 69% win rate, and together they lost $9,035.
The wins averaged $306. The losses averaged $1,442, because a losing trade had no stop and was held until the end of the day, by which time a same-day option is worth almost nothing. With losses about 5 times the size of the wins, the strategy needed to win 82% of its trades just to break even.
-$9,035net, 1 contract a trade
39live paper trades
69%made money
4 of 11sessions lost money
Running P&L by session. Red: all versions together. Grey: each version on its own. Paper trades bought at the real ask and sold at the real bid.
What is an EMA crossover strategy?
An exponential moving average (EMA) is an average of recent prices that gives more weight to the latest ones. A crossover strategy uses two of them, a fast one and a slow one. When the fast EMA crosses above the slow one, the trend has turned up and you buy; when it crosses below, you sell or go short.
It's one of the oldest and most widely taught trend-following rules in trading. The 18 and 50 period pair on a 15-minute chart is a common choice for day traders, and my backtest of it looked good enough to trade live.
The exact rules I tested
These rules were fixed before the first live trade.
Chart: SPX on 15-minute candles, regular session only. The 18-period and 50-period EMAs of the close run continuously from day to day; they are never reset in the morning.
Signal: at the close of the first 15-minute candle where the 18 EMA finishes on the other side of the 50 EMA. Crosses during a candle don't count, only at its close. 18 above 50 is a buy signal (calls); 18 below 50 is a sell signal (puts). No volume, direction or time-of-day filter.
Entry: buy the at-the-money same-day SPX option straight away at the asking price. The "discount" version instead rests a limit order 5% under the ask for up to 15 minutes, and measures its target from the original ask.
Exit: sell when the option's bid reaches +30% (or +20% for one version) of the price paid, confirmed on two checks in a row. There is no stop. Anything still open is sold at 15:55.
Late signals: a cross on a candle closing at or after 15:30 is skipped.
Results, version by version
Version
Trades
Won
Net
+30% target, bought at the ask
13
9
-$3,110
+20% target, bought at the ask
13
9
-$3,890
+30% target, bought 5% under the ask
13
9
-$2,035
All versions
39
27
-$9,035
The three versions trade the same 13 signals, so this is really 13 bets traded three ways. On 9 of them the option hit its target. On the other 4 it never got there and was sold at the end of the day for close to nothing.
Why it failed
Small wins, total losses. A winner was capped at +20-30%. A loser had no stop and rode a same-day option into the close, where it loses most of its value whatever the index does. The worst trade bought a call for $23.60 and sold it for $0.00: -$2,360 on one contract. With an average win of $306 against an average loss of $1,442, it needed to win about 82% of the time. It won 69%.
The wrong vehicle for a slow signal. A 15-minute crossover is a trend signal: in my research the move it predicts took a median of about two hours to arrive, and often went against the trade first. A same-day option bleeds value every minute it waits. The signal and the instrument work on different clocks.
The backtest was small and flattering. It covered 12 signals over six weeks of a steadily rising market, and 11 of them won. Live, 9 of 13 won, and the 4 that didn't wiped out the rest.
It didn't hold up on a second market. On QQQ, the first independent market I checked, the same crossover lost money clearly, not just by chance.
Other EMA strategies I tested
The crossover on the index alone: on SPX points, with no option, the best result came from holding to the end of the day with no profit target at all. That's a trend-following shape, positive in one rising market and unproven beyond it.
Other timeframes: the same crossover on 1-minute candles lost money; on 5-minute candles it only broke even with a one-hour trend filter, and that same filter made things worse on the 15-minute chart.
Buying the pullback to the EMA: waiting for price to dip back to the 18 EMA in an uptrend looked strong at lunchtime on SPX and SPY, but failed on QQQ, and through a real same-day option the index gain turned into a small loss per contract.
Higher highs and higher lows, then the EMA: the classic "trend structure plus pullback" entry on SPY and QQQ over nearly nine years. Every target and time exit was flat or negative, and the opposite structure earned the same.
Did I test it wrong?
Quite possibly. I tested the version of EMA crossover described above, with settings I thought were fair, and it didn't work. That doesn't prove every version fails. I may have picked the wrong settings, the wrong market or timeframe, or missed a filter that experienced traders take for granted. Many strategies also depend on judgement that a set of fixed rules can't capture.
If you trade EMA crossover and it works for you, I'd genuinely like to hear how. Tell me the chart and timeframe, what counts as a setup, any filters, where you get in, where your stop and target go, and when you get out. I'll test your version on the same data, the same way, and publish the result here, whichever way it comes out.
A version that holds the trend long enough to pay for itself, through something that doesn't decay by the hour: index futures, shares, or options with weeks left rather than hours. It would also have to work on QQQ, not just SPX.
Questions
What is an EMA crossover strategy?
Two exponential moving averages, one fast and one slow. When the fast one crosses above the slow one you buy, when it crosses below you sell. It is one of the oldest trend-following rules there is.
Why did it lose money with a 69% win rate?
Because the wins were small and the losses were total. The winners took a fixed profit of 20-30%; the losers had no stop and were held to the end of the day, where a same-day option is worth close to nothing.
Would a stop-loss have fixed it?
A tight one would have cut the winners too: in my backtest, almost half the crossovers moved 10 points against the trade before they moved 10 in its favour. The real problem is the vehicle. A trend signal that needs hours to play out doesn't suit an option that expires the same day.
Does the EMA crossover work on the index itself?
On the SPX index with no options, the crossover was positive in my backtest, but only when held for the whole trend, and the sample was one rising market.
How it was tested
Live paper trading: every trade above was taken by an engine running in real time, filled at real market bid/ask quotes on same-day SPX options, one contract each, without commissions. Nothing was added or removed afterwards.
Versions share signals: several versions trade the same setups with different exits, so the trade count overstates how many independent bets there were.
Other tests: the timeframe, pullback and trend-structure tests were backtests on historical 1-minute data, from a few months to nearly nine years depending on the test.
Still running: the engine is still trading on paper. These numbers are as of 13 Oct 2026.
This page records my own testing. It isn't financial advice. Same-day options can go to zero within minutes.
