RSI, EMA, MACD and Bollinger in a crypto bot: what the backtest showed

Every crypto bot claims its indicators "improve entry timing." The question nobody answers is: by how much, and in which market? We tested the 5 technical indicators Theta offers as an optional entry trigger — RSI, Simple Moving Average, EMA, MACD, Bollinger Bands and Stochastic — against plain DCA (buy on schedule, no filter), across 3 real market regimes.

The result isn't a leaderboard. It's a map of where each one helps and where each one loses — because that's what decides whether an entry filter is worth it for you. Backtests inform, they don't predict: past results don't guarantee future ones.

See Theta — 30 days free, no cardNo automatic charges. Cancel anytime.

⚠️ Crypto involves risk: you may lose part or all of the capital you invest. Theta is not a financial advisor and does not guarantee any return. Only invest what you can afford to lose.

Why trust it

Trade-only key, no withdrawals

The API key reads and trades — but can't withdraw. Your funds never leave your exchange account.

Never sells at a loss

No stop-loss, by design: if price drops, the bot holds and reopens later. You only realize a gain.

Free test mode

Simulate with a virtual balance, no time limit, before risking a single real cent.

Grows on its own, public plan

Half of each profit reinvests, half goes to a vault; the pilot climbs from $100 to $10,000 following a published plan. A rule, not a promise.

The scoreboard, including the losing side

We tested BTC and ETH on real 1h Binance candles, with a daily decision and a fixed budget, net of fees (0.1% per buy), across 3 market regimes: a downtrend (2022), a strong rally (2023-24) and a sideways market (2024) — always against plain DCA (buy on schedule, same pace, no indicator). Same total invested on both sides, so the comparison is fair.

Downtrend (2022): almost every indicator beats DCA — buying the dip improves the average entry price.
Strong rally (2023-24): EVERY indicator loses to DCA — waiting for a dip means buying little and late.
Sideways (2024): most gain slightly (around +1%).
Final scoreboard, out of 6 scenarios: RSI 4/6, Simple Moving Average 4/6, EMA 4/6, Stochastic 4/6, MACD 3/6, Bollinger 3/6.

RSI (Relative Strength Index)

Measures the strength of recent price action on a 0-100 scale — near 0 is oversold. Theta only opens the 1st buy when RSI hits the condition (e.g. below 30 = buy the fear).

In our backtest: beats DCA in the downtrend (+2.1% to +2.7%) and gains slightly sideways, but loses badly in a strong rally (−7% to −15%) — the filter delays entry during a continuous climb. Wins 4 of 6 scenarios.

Simple Moving Average and EMA

Both buy when price is below the average of the last N periods — EMA reacts faster because it weights recent prices more.

In our backtest, these two were the most balanced: they beat DCA in the downtrend and sideways (+0.2% to +1.2%) and lose only a little in the rally (−0.4% to −1.3%). Both win 4 of 6 — and when they lose, they lose small.

MACD

Measures the gap between a short and a long moving average — opens the buy when momentum turns up.

In our backtest, it was the most neutral: stays close to DCA in most scenarios (−0.4% to +0.7%). Wins 3 of 6, with no big swings either way.

Bollinger Bands

Opens the buy when price pierces the lower band — a statistical overreaction to the downside.

In our backtest, it had the widest swing: the best gain sideways/downtrend (+2.4% to +4.7%), but the WORST loss in a strong rally (−10% to −15%), because it buys rarely. Wins 3 of 6 — surgical, but needs a sideways market.

Stochastic (%K)

Measures where price sits within the period's min-max range — similar to RSI, different formula.

In our backtest: gains in the downtrend/sideways (~+1%), loses in the rally (−2.4% to −4.5%). Wins 4 of 6.

What this actually means

No indicator is magic, and none beats the market in a strong, continuous rally — all of them lose to plain "always buy" in that scenario, because waiting for a dip is expensive when price only goes up. Where indicators genuinely help is in downtrends and sideways markets: buying oversold improves the average entry price.

That's why, in Cycle mode, the indicator is always optional and only filters the 1st buy — never re-entries, which follow the dip/cooldown rules. You decide whether to turn the filter on, knowing where it helps and where it costs.

Frequently asked questions

Which indicator returns the most?

None wins in every scenario. Simple Moving Average and EMA were the most balanced (4/6, small losses); RSI and Stochastic also won 4/6 but with bigger losses in a rally. Bollinger has the best sideways gain and the worst rally loss. Past results don't guarantee the future.

Do technical indicators work on crypto?

It depends on the market regime. They helped in downtrends and sideways markets in our tests; in a strong continuous rally, all of them lost to plain DCA — waiting for a dip delays entry.

Does Theta force you to use an indicator?

No. It's an optional trigger that filters only the Cycle's 1st buy. Without it, the bot enters right away; with it, it waits for the indicator's condition.

Do these numbers guarantee future results?

No. Backtests inform, they don't predict. These are tests on real history (BTC+ETH, net of fees, 3 regimes), but markets change — the past doesn't repeat by decree.

See Theta — 30 days free, no card

No automatic charges. Cancel anytime.

30 days free, no card · then R$ 59/mo (from R$ 97) or R$ 490/yr (from R$ 790) · founder pricing locked until Dec 31, 2026. Outside Brazil: $12/mo (from $19) or $99/yr (from $149). Crypto (USDT): $12 = 1 month, $24 = 3 months, $79 = 1 year.