AMD's volatility was higher than Tesla's. Its drawdown was shallower.
AMD's volatility figure under Williams %R reversal, 37.81, is the highest of any ticker on this site with at least three trades. Tesla's, under the same strategy, is 24.14, well below AMD's. Tesla still drew down deeper and lost money. AMD made 308 percent.
AMD's volatility figure under Williams %R reversal is 37.81, the highest of any ticker published on this site with at least three completed trades. Tesla, traded under the exact same strategy, reads 24.14, well below AMD's and not close to the top of the list. If volatility tracked risk the way it is often assumed to, Tesla should have come through the calmer ride. It did not. Tesla's maximum drawdown, negative 61.94 percent, is deeper than AMD's, negative 40.20 percent, and Tesla's full 30-trade record lost 49.99 percent while AMD's 46-trade record gained 308.45 percent.
What volatility_pct is actually measuring
The volatility figure describes how spread out a ticker's individual trade returns were, not how deep the account went underwater and not which direction the spread leaned. AMD's 46 trades under Williams %R reversal ranged widely, a best trade of 37.77 percent against a worst of negative 26.58 percent, and 71.74 percent of them closed positive. The wide spread is real, and most of it sat on the winning side. Tesla's 30 trades were less spread out by this measure and still leaned badly: an average trade of negative 2.29 percent despite a 60 percent win rate, meaning the losing minority outweighed the winning majority, and a worst single trade of negative 37.52 percent. Volatility counts how far trades scattered from the average. It does not count whether the scatter mostly went up or mostly went down.
Why the drawdown ran deeper on the calmer-looking ticker
A maximum drawdown is a path measurement, the worst peak-to-trough decline across the whole sequence of trades, and a sequence of moderate losses stacked back to back can carve a deeper hole than a sequence of larger but more mixed results. Tesla's profit factor, gross profit divided by gross loss, is 0.725, below the 1.0 breakeven line Tenachine's guide on NR7's own risk to reward ratio already uses as the dividing line between a strategy that gives back more than it makes and one that does not. AMD's profit factor is 2.6897, comfortably above it. A ticker with more trades landing on the losing side of that line, even with a lower volatility reading, can dig a deeper drawdown than one with a wider spread of outcomes that mostly resolved in its favor.
Tenachine's guide on AMD's 308.45 percent result already covers this same trade count and win rate from a sample-size angle, checking it against Williams %R reversal's aggregate 69.13 percent win rate and finding AMD's 71.74 percent close to what the strategy does everywhere. That guide does not mention volatility or drawdown at the individual ticker level. This is the missing half of the same pair: the strategy's best raw result and one of its worst, sitting on opposite ends of the same study, with the volatility figure pointing the wrong way if read as a stand-in for risk.
This is not a claim that Tesla is a worse stock than AMD, or that Williams %R reversal is the wrong strategy for either. Both figures are backtest outcomes over one specific decade, on one specific rule, and neither ticker's result forecasts what happens next. Tenachine's guide on the study's own worst-case comparison already found a similar mismatch one level up, at the whole-strategy level rather than the single-ticker level: the strategy with the smallest average drawdown on this site was also the only one that lost money. This pair shows the same disconnect can show up inside a single study, between two tickers, using a completely different metric than drawdown itself. Volatility is a real, useful number. It answers how spread out the trades were. It was never built to answer how deep the account fell or which direction the ride went, and Tesla and AMD, run through the identical rule, are the clearest case on this site for keeping those questions separate.