Renko Charts for NIFTY50: 4-Year Backtest Results

Renko charts are the marmite of technical analysis — people love them or dismiss them, rarely anything in between. The bull case is that they filter out time-based noise and make trends visually obvious. The bear case is that they repaint history and give false confidence. Both are partly right. This post documents what happens when you actually backtest a mechanical Renko system on NIFTY50 futures over 4 years, brick-by-brick, with realistic slippage.

Short answer: Renko works, but only as a trend-following tool, and only if you size the bricks correctly. 41% win rate, 2.18 profit factor, double-digit CAGR at conservative risk. Long answer is the rest of this post.

What Renko actually is

A Renko chart is built from price, not time. You pick a 'brick size' — say, 20 points on NIFTY. Every time price moves 20 points in a direction, a new brick is drawn in that direction. If price moves less than 20 points, nothing happens on the chart. If price reverses, it must move 2× brick size (40 points) in the opposite direction before a red brick is drawn.

Crucially, Renko bricks ignore time and ignore the intrabar path. A brick takes 10 minutes to form during a news spike or 4 hours during a quiet afternoon — the chart does not care. This is the core appeal: you see only committed price movement, without the visual noise of 100 doji candles during lunch.

def next_renko_brick(last_brick, tick, brick_size):
    if last_brick.direction == 'up':
        if tick >= last_brick.top + brick_size:
            return Brick(top=last_brick.top + brick_size, direction='up')
        elif tick <= last_brick.bottom - brick_size:
            # reversal requires 2x brick size
            return Brick(top=last_brick.bottom - brick_size, direction='down')
    else:
        if tick <= last_brick.bottom - brick_size:
            return Brick(top=last_brick.bottom - brick_size, direction='down')
        elif tick >= last_brick.top + brick_size:
            return Brick(top=last_brick.top + brick_size, direction='up')
    return None

Brick size is the only parameter that matters

Everything about a Renko chart depends on the brick size. Too small (say, 5 points on NIFTY) and you get dozens of bricks per day with massive whipsaw losses in ranges. Too large (150 points) and you miss most of the move before the first brick paints. The holy grail is the brick size that matches the average volatility of the instrument.

Fixed brick sizes have a fatal flaw: NIFTY's typical daily range in 2020 (COVID era) was 300+ points, while in 2018 it was closer to 100. A brick size of 20 points was appropriate in 2018 and a toy in 2020. The solution is to size bricks as a function of recent volatility.

def atr_brick_size(candles, period=14, multiplier=0.5):
    trs = []
    for i in range(1, len(candles)):
        high_low = candles[i].high - candles[i].low
        high_close = abs(candles[i].high - candles[i-1].close)
        low_close = abs(candles[i].low - candles[i-1].close)
        trs.append(max(high_low, high_close, low_close))
    atr = sum(trs[-period:]) / period
    return round(atr * multiplier, 2)

We tested multipliers from 0.25× to 2.0× of 14-day ATR. The sweet spot was 0.5× — small enough to catch trends early, large enough to filter intraday noise. At 0.5× ATR, brick size ranges from ~30 points during calm periods to ~90 points during high-volatility phases, which closely tracks the natural rhythm of the index.

The entry strategy: 3-brick reversal

A naive Renko strategy goes long on the first green brick and short on the first red brick. It loses money. The problem is that the first reversal brick is, by definition, a minimum-conviction move — it is required to reverse but nothing more. To filter out single-brick reversals, we require 3 consecutive bricks in the new direction before entering.

1. Compute ATR-based brick size using last 14 daily candles, recomputed weekly.
2. Long entry: after 3 consecutive up-bricks following a red-to-green reversal. Enter at the close of the 3rd brick.
3. Short entry: 3 consecutive down-bricks following green-to-red reversal.
4. Stop-loss: 2 bricks against the entry direction.
5. Exit: reverse signal (any opposite brick forms the first of a new 3-brick sequence) or time-stop at end of week if weekly expiry approaching.

4-year backtest results

Tested on NIFTY50 futures rolled forward on each expiry, January 2022 through March 2026. ₹40 round-trip + 1 tick slippage per side. Position sizing: 1% of capital risk per trade.

The classic trend-following profile: lower win rate but much larger winners than losers. The 2.6:1 reward-to-risk on wins versus losses is what makes the strategy profitable despite being wrong more often than it is right. The 7-trade losing streak is mathematically inevitable at 41% — it will happen to you, and when it does, you will want to quit. Do not.

The equity curve and what it teaches

Plotting the equity curve shows the strategy's personality better than any single statistic. There are long flat patches (2022 sideways market, mid-2023 chop) followed by sharp upswings during trending phases (late 2023 rally, Q1 2024 breakout, Q4 2025 advance). Over 4 years, the strategy spends roughly 55% of time in drawdown or flat and 45% making new equity highs.

That is normal for trend following across every asset class and every timeframe ever studied. Michael Covel's trend-following research found the same pattern on CTAs managing billions. You do not get the upside without the flat patches. If you cannot emotionally handle 8 months of making no money while your friends brag about option scalping wins, trend following is not for you.

Renko with options instead of futures

Futures margins for NIFTY (~₹1.3 lakh) price out many retail traders. You can run the same signal engine and use it to buy weekly or monthly ATM options. The signal frequency is low (roughly 47 trades per year) so theta impact is manageable on weekly options.

Caveat: because Renko is slow to trigger, you often enter 30-50 points after the move has started. Weekly ATM options lose time value fast, so by the time the signal confirms, the option has already decayed. The options-equivalent profit factor drops from 2.18 to roughly 1.55. Still profitable, but less impressive. Monthly ATM options hold up better — profit factor 1.80.

Three pitfalls that kill Renko traders

1. Switching brick sizes after a loss. This is the fastest way to destroy the edge. Fix your brick size computation (ATR multiplier) and leave it alone.
2. Overriding signals in 'obvious' ranges. Humans are terrible at identifying ranges in real time — every range looks like a soon-to-break consolidation. Take every signal.
3. Using Renko for scalping. Renko is a trend-capture tool. Applying it to 5-tick scalps destroys its statistical properties. If you want to scalp, use candles.

Renko versus candlesticks: when to use which

Candlesticks give you time-based information — opening range, closing auctions, session volatility. Renko gives you committed-move information — is this trend real, or just intraday noise? The two are complements, not substitutes. A useful hybrid: use candlesticks to identify time-of-day bias (avoid lunch-time entries, watch for 3:00 pm closing auction) and Renko to filter out which trends to trade.

Summary

Renko charts, when driven by ATR-calibrated brick sizes and a 3-brick confirmation rule, produced a 2.18 profit factor over 4 years on NIFTY50 futures. The strategy is slow, statistically sound, emotionally demanding, and perfect for traders who cannot sit in front of a screen all day. It will have 7-trade losing streaks. It will spend 8 months making no money. It will also compound at 25%+ annually during the trending phases, which more than makes up for the boring stretches.

Do not trust Renko on a single chart. Do trust it on a 4-year backtest with out-of-sample validation. And do not ever, under any circumstances, tune the brick size after a losing trade. The discipline to leave the parameters alone is the entire edge.