Faber Timing Model

How this strategy trades, what it scored, and the measurements the score was read from: every symbol and timeframe, on real and resampled history.

At A Glance

60/ 100
Total Score
8of 52
Overall Rank
8of 28
In Trend
5,566
Trades Taken

Across 24 runs of real history, one for every market and bar size it trades, this strategy scored highest on SOLUSDT · 4h (88 / 100) and lowest on XRPUSDT · 4h (26 / 100).

What A $100.00 Stake Became

Score
Total Score Mean Of Real And Resampled 60 / 100
Real Data 64 / 100
Resampled Data 55 / 100
Overall Rank Top 16% 8 of 52
Rank In Trend Top 29% 8 of 28
By Metric
Beats Hold Real 65 · Resampled 67 66 / 100
Risk Adjusted Real 56 · Resampled 44 50 / 100
Profitability Real 79 · Resampled 59 69 / 100
Win Rate Real 30 · Resampled 28 29 / 100
Fee Efficiency Real 86 · Resampled 86 86 / 100

Where It Works

How It Trades

  1. Compute a 10-month simple moving average of the close: 210 bars, treating every 21 bars as one month.
  2. Check the strategy only once a month, on every 21st bar; between reviews, nothing the market does matters. This monthly cadence is Faber's key idea: it ignores intra-month noise entirely and is what keeps the trade count tiny.
  3. At a review: if the close is above the 10-month average, be in the market; if below, be out. Buy or sell only when the answer changes from the previous review.

Example Strategy

Pine Script

Paste it into TradingView’s Pine Editor to run these rules on any chart.

//@version=6
// Faber Timing Model — sort.finance. Fills at the signal bar's close, 0.02% fee a side,
// whole account per trade, as the backtest. History and indicator warm-up
// differ from the site's, so expect close to the same trades, not identical.
// The review dates count from the first bar on the chart, and TradingView's
// history starts somewhere other than the backtest's, so the reviews here
// fall on different bars from the site's.
strategy("Faber Timing Model", overlay=true,
     initial_capital=100, default_qty_type=strategy.percent_of_equity,
     default_qty_value=100, commission_type=strategy.commission.percent,
     commission_value=0.02, process_orders_on_close=true, pyramiding=0)

// The bar the current position opened on, and the bar the last one closed on.
var int entryBar = na
var int exitBar = -1

// ── Signals ──────────────────────────────────────────────────────────────
average = ta.sma(close, 210)
// Reviewed once a "month" — every 21st bar from the 210th on — and
// only a change between two reviews trades.
review = bar_index >= 209 and (bar_index - 209) % 21 == 0
var bool reviewed = false
var bool wasAbove = false
enterNow = false
exitNow = false
if review
    nowAbove = close > average
    if reviewed
        enterNow := nowAbove and not wasAbove
        exitNow := not nowAbove and wasAbove
    wasAbove := nowAbove
    reviewed := true

// ── Orders ───────────────────────────────────────────────────────────────
// Open on the first bar the entry holds while flat; close on the first
// later bar the exit holds; never reopen on the bar a position closed.
inTrade = strategy.position_size != 0
if inTrade and bar_index > entryBar and exitNow
    strategy.close("Long")
    exitBar := bar_index
else if not inTrade and bar_index > exitBar and enterNow
    strategy.entry("Long", strategy.long)
    entryBar := bar_index

// ── Drawing ──────────────────────────────────────────────────────────────
plot(average, "SMA 210", color=#4d8dff)
Show

Total Score

64 55 / 100

Score = 0.25 Sbeats_hold + 0.35 Srisk + 0.20 Sprofit + 0.10 Swin + 0.10 Sfee

One 0–100 number per symbol × timeframe cell. Every strategy is run on every cell, on both real and resampled price history, and the total beside its name is this number averaged over all of them.

Beats Hold

25% of the total

65 67 / 100

Sbeats_hold = clamp0100(50 + 25 log2EnetEhold)

Compares final net equity to what simply holding the asset would have returned over the same window. 50 is break-even; every doubling versus buy-and-hold adds 25 points.

Read From Net Equity · Price Change %

Net Equity

The account after fees, trade by trade.

Price Change %

What holding the asset made over the same stretch: the hold the account is measured against.

Risk Adjusted

35% of the total

56 44 / 100

Srisk = clamp0100(55 · Sharpe + 5)

A linear read on the rolling annualized Sharpe ratio at the last trade: 0 Sharpe scores 5, roughly 1.7 Sharpe maxes out the scale.

Read From Rolling Sharpe

Rolling Sharpe

The account's annualized Sharpe ratio at each trade's close.

Profitability

20% of the total

79 59 / 100

Sprofit = clamp0100(20 log2EnetE0)

How many times the starting cash multiplied, on a log scale: each doubling of capital is worth 20 points.

Read From Net Equity · Cumulative P&L %

Net Equity

The account after fees, against the cash it started with.

Cumulative P&L %

Not in the formula: every trade's return added up rather than compounded.

Win Rate

10% of the total

30 28 / 100

Swin = clamp0100(2.5 (W% − 20))

The share of trades that closed profitable, rescaled so a 20% win rate scores 0 and a 60% win rate maxes out the scale.

Read From Win Rate % · Trades

Win Rate %

The share of trades so far that closed in profit.

Trades

How many trades that share is taken over.

Fee Efficiency

10% of the total

86 86 / 100

Sfee = clamp0100(Enet/Egross − 0.40.6 × 100)

The share of the fee-free (gross) result that survives after fees: 60% survival scores 0, keeping the full gross result scores 100.

Read From Equity, Net And Gross · Cumulative Fees

Equity, Net And Gross

The account with the fees taken and, fainter, without them.

Cumulative Fees

Not in the formula: what the exchange took, which is the gap between net and gross above.