This report analyzes a short call option strategy on Apple Inc. (AAPL) using historical data from Feb 2023 through Jan 2026 (approximately 751 trading days). The strategy involves selling one out-of-the-money AAPL call option with roughly 17 delta and about 45 days to expiration (DTE), with two management variants:
- Close at 15 DTE: Sell a ~45 DTE call and close the position once the option reaches 15 DTE, remaining out of the market until the next ~45 DTE entry point.
- Roll at 15 DTE: Sell a ~45 DTE call and when it reaches 15 DTE, roll the position by closing the expiring short call and immediately opening a new ~45 DTE call (maintaining continuous exposure).
Performance Summary: Both strategy variants produced net losses over the tested period, but the continuous roll approach significantly reduced the magnitude of losses and drawdowns compared to the close-only approach. Key performance metrics include:
- Total P&L: –$3,939 (Close) vs. –$237 (Roll) – The roll strategy nearly broke even, losing only a fraction of the close strategy’s losses.
- Win Rate: 61.3% (Close) vs. 65.8% (Roll) – Both had a majority of winning trades, with the roll variant slightly higher.
- Max Drawdown: –$4,418 (Close) vs. –$1,448 (Roll) – The worst peak-to-valley equity decline was much smaller for the roll strategy.
- Sharpe Ratio: –0.50 (Close) vs. –0.07 (Roll) – Both Sharpe ratios were negative (no risk-adjusted outperformance), but the roll’s Sharpe was near zero, reflecting its much smaller loss.
In summary, continuously rolling the short call position at 15 DTE spreads risk across more trades and time periods, resulting in smaller cumulative losses and shallower drawdowns than the periodic close-only strategy. However, the strategy in both forms suffered from asymmetric risk: many small profits were eventually outweighed by a few significant losses. In the volatility regime analysis, nearly all the losses occurred when implied volatility was low or average, whereas initiating trades during high-IV periods was notably more successful. This hints that filtering trades by market volatility could materially improve outcomes. Overall, the backtest underscores both the potential and the significant tail risks of short call strategies, informing potential refinements to enhance risk-adjusted returns.
Strategy Mechanics and Methodology
- Underlying and Data: The backtest sells call options on AAPL. Historical option data was sourced from Massive.com’s option aggregates, and all Greeks and implied volatilities (IV) were computed in-house using Black–Scholes. The study covers 2023-02-01 to 2026-01-29, with trades recorded throughout this period.
- Entry Rule: On each eligible entry date, sell one call option ~45 DTE with ≈17 delta. To simulate realistic execution, the entry fill is assumed at the midpoint minus half the bid-ask spread (i.e. paying a 50% of half-spread “slippage” on entry).
- Exit and Roll Rules: If holding a position, when the option’s DTE falls to 15 days, buy back the call to close the position. The exit fill is assumed at the midpoint plus half the spread (again accounting for slippage in closing).
- In the Close at 15 DTE variant, the strategy then stays in cash (no position) until the next suitable ~45 DTE entry opportunity arises (typically the next expiration cycle).
- In the Roll at 15 DTE variant, on the same day, one position is closed at 15 DTE, a new ~45 DTE, ~17Δ call is immediately sold to keep the short call position continuously active. This rolling mechanism increases the number of trades and maintains nearly continuous market exposure.
- Trade Filters and Execution Assumptions: To avoid illiquid options, any candidate option with a midpoint price below $0.05 or with an excessive bid-ask spread (greater than 50% of the option’s mid price) is skipped. Each trade incurs a commission of $0.65 per contract per side (opening and closing, total ~$1.30 round-trip). All trades are one contract short, and no position sizing or compounding is applied (P&L is tracked in dollars per contract).
- Volatility Regime Tagging: Each trade is labeled as “High IV” or “Low IV” regime based on the entry date’s implied volatility level. The criterion uses a 252-trading-day lookback: it computed the median implied volatility of AAPL options on the entry day, and compared it to the 80th percentile of the past 252 days’ median IVs. If the day’s IV is at or above this 80th-percentile threshold, that trade is tagged as a High IV entry; otherwise, it’s a “not high” (normal/low IV) entry. This enables performance comparisons across different volatility environments. In our sample, the high-IV threshold corresponds to an implied volatility of ~27% (for context, AAPL’s median IV was typically lower than this in 2023–2025).
The backtest was conducted sequentially for both strategy variations. The Close 15 DTE strategy executed 31 trades over the 3 years, whereas the Roll 15 DTE strategy executed 38 trades (rolling more frequently resulted in a higher total number of trades). Trades generally lasted around one month each (enter ~45 DTE, exit at 15 DTE ≈ , 30 days held). Since the roll strategy opens a new trade on the same day as closing the previous one, it achieves near-continuous exposure. In contrast, the close strategy was out of the market for several days or weeks between some trades (while waiting for the next 45-DTE entry point). We next examine the performance outcomes and risk characteristics of these two approaches.
Performance Analysis
Figure 1: Cumulative P&L (equity) curves for the short-call strategy variants over time, starting at $0. Orange shows the “Close at 15 DTE” strategy, and red shows “Roll at 15 DTE.” The roll strategy clearly outperformed in this backtest, ending near breakeven (slightly negative), whereas the close strategy accumulated a significant net loss. The roll approach’s equity line is higher throughout most of the period, indicating smaller drawdowns and quicker recovery after losses.
Overall P&L Trajectory: As shown in Figure 1, both strategies exhibited a volatile equity curve, with generally declining performance punctuated by occasional rebounds. The Close-at-15-DTE variant (orange line) ended with approximately –$3,939 in total profit, meaning a net loss over the period. In contrast, the Roll-at-15-DTE variant (red line) finished around -$237, nearly breaking even. Notably, the roll strategy’s equity stayed higher than the close strategy’s equity for most of the backtest. Early in the test (2023), both strategies saw slight gains and losses without a clear trend, but by mid-2024, the close strategy began to accumulate a substantial drawdown. The roll strategy, with its continuous exposure, also incurred losses, but to a much lesser extent.
Roll vs. Close – Why the Difference? Rolling at 15 DTE improved outcomes by re-entering trades sooner and increasing the number of trades, thereby distributing risk across more independent bets. The close-only approach often remained in cash during the same periods when the roll strategy could initiate new positions (some of which proved profitable). In other words, the roll strategy captured additional premium opportunities that the close strategy missed. This resulted in the roll variant having 7 more trades than the close variant (38 vs 31), and those extra trades helped offset losses by yielding additional winners. The win rate was slightly higher for the roll strategy (66% vs 61%), and importantly, the average loss was much smaller (≈–$250 for roll vs –$536 for close). This suggests that the worst outcomes in the close strategy had a disproportionate impact. By contrast, the roll strategy’s continuous exposure did not translate into larger losses; in fact, it mitigated significant losses by resetting positions monthly and not leaving the strategy idle during potentially favorable periods and in essence, rolling provided a form of time diversification: more trades, each with smaller loss potential, rather than fewer trades where one loss can severely damage the cumulative P&L.
Core Performance Metrics: Both strategy variants had more winning trades than losing trades, but the distribution of gains and losses resulted in negative overall returns. The profit factor (ratio of total profit from winning trades to total loss from losing trades) was a dismal 0.39 for the close strategy, meaning losses outweighed gains by a large margin. The roll strategy’s profit factor was 0.93 – still below 1.0, but much closer to breakeven, reflecting a near balance between premiums earned and losses paid out. Neither strategy produced a positive Sharpe ratio, as both resulted in net losses. The roll variant’s Sharpe (≈ –0.07) was almost flat, considerably better than the close variant’s Sharpe (~ –0.50), again underscoring that rolling reduced volatility of outcomes and nearly achieved a net-zero return (albeit slightly negative). In practical terms, an investor implementing the close strategy would have lost approximately $3,939 per contract sold over three years. In contrast, the roll strategy would have broken even (a $237 loss, which is within the noise of transaction costs over that period).
Drawdowns: A critical difference between the approaches was the depth of drawdown. The worst peak-to-trough equity drawdown for the close strategy reached –$4,418, meaning that at one point the strategy was down more than $4.4k from its high. By contrast, the maximum drawdown on the roll strategy was –$1,448, about one-third of the close strategy’s. The close strategy’s equity curve (Figure 1, orange) shows a sharp drop in mid-2025 that took a long time to recover (indeed, it never fully recovered, ending near its low). The roll strategy’s drawdowns were gentler and had quicker partial recoveries. Smaller drawdowns not only indicate lower risk but also make it psychologically easier for a trader to stick with the strategy. The much shallower drawdown of the roll variant is a key point in its favor, resulting directly from avoiding long idle periods and thus having more chances to recover after a loss.
Risk/Reward Profile and Tail Risks
The short 17Δ call strategy inherently has a skewed risk/reward profile: it generates frequent small gains and occasional significant losses. This is evident in the distribution of trade outcomes for our backtest.
Figure 2: Distribution of individual trade P&L for each strategy variant (per-contract). Left: “Close at 15 DTE” trades (orange); Right: “Roll at 15 DTE” trades (red). Both histograms show the count of trades at various P&L levels. The close strategy exhibits a long left tail – a few trades incurred huge losses (far left bins around –$1500 to –$1900), while the majority of outcomes were small gains or modest losses near $0 to +$200. The roll strategy’s loss tail is shorter, with its worst loss below –$1000, and most trades clustered near breakeven or small profits. This highlights the strategy’s asymmetric payoff: high-probability small wins versus low-probability significant losses.
As Figure 2 illustrates, most trades in both variants returned modest profits or small losses, clustered around the breakeven mark. A win rate above 60% indicates that a majority of positions expired or were closed profitably (the short call premium was earned and not fully offset). However, the losing trades, while fewer in number, were often much larger in magnitude. In the close strategy (orange histogram), there are a few extreme loss outliers: in fact, 3 trades each lost more than $900, including one outlier approaching a $1,900 loss. These significant losses dwarf the typical win size (the average winning trade was only $130 for close). The roll strategy (red histogram) also incurred losses, but its worst loss was approximately $939, and only one trade exceeded $500 in losses. The roll variant’s average loss was roughly $250, less than half that of the close variant. By curtailing tail losses, the roll approach produced a distribution that, although still negatively skewed, was less extreme.
To put this in context, the worst trade in the close strategy occurred when AAPL’s stock price rallied strongly against the short call. In that July to September 2025 trade, a 220 strike call was sold for about $1.99 and had to be repurchased for $20.80 as it went deep in-the-money, resulting in a –$1,894 loss on one contract. This single loss was equivalent to erasing the profit from roughly 15 typical winning trades. The roll strategy’s worst case was a loss of about $939 on a 200 strike call in mid-2024; painful, but only about half as severe as the close strategy’s worst loss. Both strategies faced a few such significant adverse moves when AAPL’s price climbed significantly above the strike before exit.
The risk profile here is typical for short calls (or any short option premium strategy with limited upside and substantial downside). The strategy tends to have a high probability of small wins – for instance, a 17Δ call has roughly an ~83% probability of expiring out of the money (ignoring early closes), which aligns with the >60% win rate observed – but when it loses, the losses can be open-ended. Without owning the underlying stock (i.e., these are naked calls, not covered calls), upside moves in AAPL can theoretically produce unlimited loss. In practice, losses are constrained by exiting at 15 DTE rather than holding to expiration, but, as we saw, even within ~30 days, a significant upward move can materially increase the call’s price, causing substantial losses.
The max drawdowns discussed earlier reflect this dynamic: the close strategy’s –$4.4k max drawdown was primarily driven by just a couple of those tail losses occurring without enough profits between them to cushion the blow. The roll strategy, by virtue of more frequent, smaller trades, avoided compounding losses and even generated profitable trades between loss events, thereby reducing drawdown severity.
Risk management for such a strategy is paramount. This backtest did not employ any stop-loss or hedging beyond the time-based exit rule (15 DTE). The data demonstrate that tail-risk events (large stock rallies) occurred and were the primary source of the strategy’s losses. An investor considering this strategy should be aware that, despite a benign win rate and steady premium income most of the time, the strategy is highly vulnerable to sudden upward moves in the underlying stock. Mitigating these risks – whether through position sizing, stop-loss rules, or by converting to a spread (buying a further OTM call as protection) – could be explored to prevent singular trades from devastating the portfolio.
Impact of Volatility Regime
A notable insight from the backtest is the stark difference in performance between high- and low-implied-volatility environments at the time of entry. The strategy was analyzed under these regimes to determine whether selling calls was more favorable when IV was elevated (indicating that the options were relatively expensive, presumably pricing in more risk).
Out of all trades in the close strategy, only 3 trades occurred on “High IV” entry days (by our 80th-percentile IV definition), and the remaining 28 trades were in more normal/low IV conditions. Despite the small sample, the results are striking: every high-IV trade was profitable. Those 3 high-IV entries yielded a total of $552 (average $184 per entry) with a 100% win rate. In contrast, the 28 trades in lower-IV conditions incurred a combined loss of $4,492, with a win rate of 57%. In other words, all of the net loss came from trades initiated when implied volatility was not exceptionally high.
Figure 3: Performance by volatility regime (Close-at-15-DTE strategy). The left chart shows the total P&L from trades entered during High IV days vs Low/Normal IV days. The right chart shows the win rate in each regime. The strategy’s entire profit came from the few high-IV trades (+$552 total, 100% win rate), whereas trades in low-IV environments lost a total of –$4,492 with a win rate below 60%. High implied volatility at entry appears to have been a favorable condition for this short call strategy.
Figure 3 summarizes the regime performance. Selling calls during high-volatility periods not only yielded positive total profits in our test, but did so with an unblemished win record (again, only three instances, so this 100% is not statistically robust, but it’s telling). Meanwhile, the considerable aggregate loss arose from trades in the low-IV regime. Intuitively, this makes sense: in higher-vol markets, call options are more richly priced (higher premiums received), and perhaps AAPL’s subsequent moves were not as extreme as the options market feared (volatility often mean-reverts). In lower-volatility regimes, the premium collected was smaller and may not have adequately compensated for the stock’s bullish drift or for surprise events. It’s possible that during low-IV periods, investor complacency was high and the stock trended upward, leading short calls to lose money even without dramatic volatility spikes. Also, when starting IV is low, an unexpected volatility surge (for instance, due to an earnings surprise or market shock) can severely harm short option positions, whereas when starting IV is high, it often declines (volatility crush) or, at least, the bar for further surprise is higher.
Another way to view it: if one had only taken trades when IV was high, the strategy would have avoided the worst losses and been profitable. A hypothetical “High-IV-only” version of the close strategy (taking just those 3 trades) ends with +$552, versus an “Low-IV-only” version (taking the other 28 trades) ending at –$4,492. Of course, in reality, one wouldn’t know future volatility regime shifts with certainty, but this finding suggests that imposing an IV filter could be a beneficial risk-management tool. It might have kept the trader out of the market during extended low-vol rallies (such as steady bull runs) when short calls perform poorly, and only engaged during more turbulent or fearful times when the option premium is high. The risk/reward for selling calls is better skewed. Meanwhile, all trades that started with IV above our high-vol threshold (dashed line) ended up profitable. While the sample is limited, it aligns with the notion that high IV can indicate a better reward-to-risk setup for selling calls – the premium cushion is larger. The stock would have to make an especially outsized move to inflict a loss. In low-IV regimes, that cushion is thin, and even routine uptrends can break through the strike price.
It is important to note that “high IV” here is defined relative to recent history (80th percentile of the past year’s IV levels for AAPL). In absolute terms, AAPL’s IV during 2023–2025 was not extremely high (roughly mid-20s percentage); it was not crisis-level volatility, but it was higher than typical for the period. Yet even that relative increase in implied vol made a substantial difference in outcomes. This suggests a practical takeaway: a trader might improve this strategy by being selective – for instance, only deploying the short call trade when the VIX or AAPL’s own IV is elevated, and staying on the sidelines during complacent, low-vol markets.
Conclusion and Takeaways
This deep-dive backtest highlights both the attractions and pitfalls of a short call options strategy on AAPL. On one hand, the strategy produces steady income most of the time (over 60% of trades were winners), and the rolling management approach demonstrated that staying consistently engaged can smooth out returns and reduce drawdowns. The roll-at-15-DTE variant outperformed the simple close-and-wait variant, reducing total losses substantially and improving the win rate and return stability. This was achieved by increasing the number of trades and maintaining continuous exposure, thereby enabling the strategy to capitalize on more opportunities and recover more quickly from losses. For a trader considering short calls, this suggests that an active rolling schedule may be superior to a set-and-forget approach in terms of balancing risk and reward.
On the other hand, the strategy in both forms incurred a net loss over this period, underscoring that shorting calls on a strongly trending stock like AAPL can be hazardous. The negative overall P&L and Sharpe ratios were driven by a handful of tail-risk events where AAPL’s price rallied strongly, causing significant losses that wiped out dozens of small gains. The risk/reward is asymmetrically skewed against the trader: one bad trade can negate months of profits. This underscores the need for risk management. Potential improvements to explore include:
- Volatility-based trade filtering: As shown, focusing on high implied volatility conditions dramatically improved performance in this backtest. An investor might only sell the 17Δ call when IV is above a certain threshold (e.g., 80th percentile of recent IV), aiming to avoid low-premium environments that don’t compensate for the risk.
- Position limits or stop-losses: Implement rules to cut losses if a trade goes deeply against you before the 15 DTE mark. For example, if the call’s price doubles or the underlying breaches a certain distance above the strike, one might close early to prevent an outsized loss. This could potentially reduce the severity of the worst trades (albeit at the cost of sometimes cutting off a trade that might have reverted).
- Spread structure or hedging: Convert the naked call into a bear call spread by buying a further out-of-the-money call for protection. This will cap the maximum loss on any given trade (in exchange for giving up some premium). While this will reduce profit per trade, it can significantly mitigate tail risk and thus improve the strategy’s risk-adjusted returns. Another approach could be to trade a covered call (selling calls against a long stock position) to offset losses with stock gains, though that changes the trade’s profile significantly.
- Underlying selection and trend: Consider that AAPL had a general uptrend in the period studied (as many large tech stocks did), which is a headwind for any short call strategy. Applying this strategy to a different underlying or during a more range-bound market might produce more favorable results. Diversifying across multiple uncorrelated underlyings could also spread risk.
- Premium intake vs. strike selection: The choice of ~17 delta (roughly one standard deviation OTM) was somewhat conservative in terms of probability of profit. One might test whether selling a higher delta (closer to the money) for a higher premium, or a lower delta for a higher win rate, improves expected return. However, higher delta short calls would risk even larger losses if wrong, so any such adjustment should be carefully evaluated.
In conclusion, the short 17Δ call strategy on AAPL is an example of an income strategy with hidden tail risks. Our backtest informed us about how different management tactics (rolling versus closing) and market conditions (volatility regimes) affect performance. The rolling strategy proved superior in this study, nearly turning a losing strategy into a breakeven one by simply keeping the position consistently deployed. Still, neither variant would be considered successful without further risk mitigation, given the negative returns and large drawdowns observed. For sophisticated traders, the key takeaways are to manage the risk proactively – by timing the strategy in favorable conditions, by cutting losses or capping risk, and by understanding that short calls, especially on a rising stock, can impose significant losses sporadically. With prudent adjustments, one might capture the appealing high win-rate income of short calls while controlling the damage from the occasional turbulent episodes when AAPL’s price surges. This research provides a foundation for such improvements: a high-IV entry filter and a disciplined rolling schedule already appear to be steps in the right direction for making the strategy more robust.
