Dialing in on TSLA covered calls

Fifty Ways to Trade an Option

The covered call is the simplest options trade.

Brokers allow the overlay in a cash account because the risks (to them) are no worse than customers owning equities. The risk to the customer is much greater.

The setup is pitched as a win-win. If stock goes down you make a little bit of premium to buffer the drop. If stock goes up you also collect that cash, and get to sell the shares a little bit higher.

If it was that easy, you’d simply take the monthly premiums, multiply by twelve, and add 10-50% points a year to your underlying investments. The below is the arithmetic done for the top 25 most liquid names, assuming you sell the 30 delta call every month for a year. Seems pretty tempting right?

What lies beneath those premiums is a hefty dose of risk. And as my friend Kris Abdelmessih says, every options trade is a volatility trade. The risk premium that you collect on the trade is in exchange for giving away the right tail of returns, and locking yourself into timing risk around expiration.

A covered call is not income, rent, or yield. That’s probably not surprising to most of you. What was surprising to me was just how instructive it could be to decompose the results of regularly trading this strategy.

What started as a look at how to outperform the systematic overlay ETFs, grew into a full blown analysis of how, where, and why you make or don’t make money with a covered call strategy. Of course Kris and I picked the most obnoxious stock possible to demonstrate this - Tesla Inc.

For six and a half years, TSLA has been on a wild ride. It’s up well over 1000% during this time period. But it’s had major drawdowns - in 2025 stock has already been cut in half, added back 50%, and it’s not even June.

The lessons from this are enough to fill over 90 minutes of air time, and that was cutting it tight. I’d encourage you to watch the full bit here.

The core strategy here was replicating a systematic overlay that sold calls every month on the underlying. Even if that sounds 101, there are plenty of implementation details. We chose the call closest to 25 delta, and only traded monthly expirations.

The calls were closed every month rather than letting shares get called away. The impact on that is meaningful. As you’d expect from a (well priced) 25 delta call, about 71% of the time they expire worthless, but in the other cases the stock position dwindles over time with rebuying at a higher price. The flip side demands quite a bit of cash to cover all the shorts.

In both cases the effect of rolling your position is to lock in the current market price. If TLSA closes at $320 and you’re short the $300 call, the $20+ you pay to close that position is out the door. If/when stock drops again, the paper gains on equity aren’t there to offset the real cash needed to close the options positions.

This opens the door to the most important lesson for covered call traders - path dependence really matters. Whether or not volatility is fairly priced (more on that later) is a consideration distinct from the directional move of the underlying.

Both within a monthly cycle and over the course of 72 distinct trades, we see how that path impacts a strategy. Expiration is the double edged sword of options. It’s what gives options a premium, but also what creates the rebalance timing risk.

Zooming out, “when” the returns happen over the course of the study has a dramatic effect on the amount of risk that you’re taking over time. While the price on the screen today looks similar (~$270 start vs. $245 at March expo), two stock splits put 15x more shares into circulation, making your risk 12x as big. That makes later trades and results significantly more impactful on the overall result.

The plummeting lines from January to March 2025 deal a significant blow to overwriters and stock holders. (The above assumes 10,000 share initial purchase). After three final disastrous months, the overwrite has only achieved a 1.28x return in the period, while the equity holder is up 11.4x. If we ended the sample in January, the call seller would only be trailing by half.

Practically this means we need to also analyze the results from a constant position sizing perspective. If 3 of 72 trades wipe out your stack, something’s off with sizing. All but the most aggressive fanatics would be rebalancing their TSLA shares with respect to the overall portfolio.

So how do we improve the results here?

Seeing all that bleeding red Pnl, a good question to ask would be if options are fairly priced. We saw a hint of that with the high level win/loss stats - 25 delta targets expired in the money about 29% of the time, suggesting the delta revealed by the options implied volatility was fairly accurate.

The option greeks are outputs of the pricing model - sensitivities to changes in underlying price, volatility or rates. To isolate how well the implied volatility matches realized volatility, we simulate delta hedging the options on a daily basis.

The net Pnl (blue) from that strategy is $3 per $100 of risk. Meaning over the course the entire study, your annual return was only 47bps. That suggests just a sliver of the volatility risk premium paid to the seller, but well less than the risk free rate. Those are very well priced options.

Fair prices come from efficient markets, and we can see that TSLA has gotten more efficient over time. Much of the PnL from hedging came early in the period, while the blue line trends sideways for the latter half of the study. Plenty of volume and activity in a highly competitive marketplace drive spreads tight and prices fair.

If overall the options are fairly priced, perhaps there are some indicators or signals we can use to determine when to trade?

The three basic volatility indicators we reviewed were Implied Volatility, Skew, and Volatility Risk Premium (VPR). The first two are forward looking, derived from market prices. VRP is historical, comparing what implied volatility was in the past to what it most recently realized.

We use the relative level of these indicators to determine whether or not to trade. Percentile rank is compared on two time frames, 30 days and 1 year. For IV and VRP we look for high values, e.g. when options are priced more expensive than usual, or there has been a recent trend of implied trading over historical. With skew we are looking for “low” values of the 25 delta put minus 25 delta call, because that would indicate greater degrees of call skew.

Using longer term indicators has the advantage of a longer and more contextualized history. It can also put a strategy out of market for a long time if there is a regime shift. If you’re looking to sell high volatility opportunities, and vol has a paradigm shift lower, there won’t be much that looks expensive. Shorter term indicators will be more reactive, but can be susceptible to noise in the sample.

Every indicator improves results over the baseline strategy, but not for particularly interesting reasons. Most of the improvement comes from reducing the trade count. If you’re looking for all the indicators to green light a trade, there are only 11/17 (30day/1 Year lookback) opportunities out of 72. It’s tough to call that an overwriting strategy.

Skew seems to be the best performing indicator, while VRP is the worst. A qualitative explanation for that would be skew indicating call bids comes from participants who prefer the optionality of a call to the stock itself. When things are relatively high vol, the chances of a big down move skew the risk/reward to owning calls. Selling into that was favorable not because of expensive options, but because of resultant price distributions to the downside.

Traders of covered calls must care about both the direction of the underlying and the price of the option. It is a long delta strategy, but with greeks that evolve at the extremes. As stock moves through the upside strike, your long stock short 25 delta call goes from long 75 deltas to flat. On the downside the option loses deltas and the position becomes 100% long if we move far enough away.

If you’re delta hedging this on a daily basis, it will highlight sometimes significant amounts of “unexplained” PnL. The greeks you hedge from the options pricing model are a snapshot in time at a specific price. Those change as the underlying makes a big move. Vega goes away as you move further from the strike, leaving you not quite as short vol as you may have hoped.

Layering options into a stock position introduces a number of different complexities. Not only is there the practical management of trading and rolling every month (every day if you’re delta hedging), a different PnL path will emerge based on when and how the underlying moves. Heading into expiration or as the position size increases, you’ll see single day movements have a much greater impact, and more unexplained - or unhedged - results.

More complex signals or more dynamic hedging could possibly improve the results here. There may even be indicators from the options markets that improve equity trading results. For example with skew, high call skew in the short term suggests calls finish out of the money more often than usual. Selling a call isn’t necessarily the best way to take advantage of that dynamic.

Whether you’re an overlay investor looking at stock + option returns, or parsing the individual greek sensitivities, there’s a lot to learn from how the simple covered call trade behaves.

A big shout out to Kris for working through this with me. If you want to see more details on PnL attribution, check out the new features over at Moontower AI.