Tesla TSLA stock prediction: $500 bull vs $250 bear

Tesla's December options board prices a 37% rally and a 31% collapse as very nearly the same bet.

The $500 call, 37.4% above Friday's close, carries a risk-neutral probability of about 10.5%. The $250 put, 31.3% below it, prices at about 10.8%. Both are heavily owned – 17,062 and 11,341 contracts of open interest on the 18 December 2026 expiry.

For most large-cap equities the downside tail is priced meaningfully fatter than the upside one. In Tesla's case the two tails are, to within a third of a percentage point, the same size. That symmetry is the single most useful input into any TSLA price prediction, and it is not what the volatility surface of a normal $1.4tn company looks like.

The reason is that Tesla's implied-volatility curve is a smile that leans right. At-the-money implied volatility for December is 45.1%, per Cboe delayed quotes. The $200 put – a 45% crash – trades at 58.0%, a 12.9-point premium. The $800 call trades at 62.8% and the $990 call at 70.7%, a 25.6-point premium over at-the-money. The market charges more to insure Tesla's upside than its downside.

Tesla closed Friday 21 August at $362.86, up 5.14% after Nevada regulators cleared a permit for up to 5,000 Cybercab robotaxis in Las Vegas. The stock remains 25.9% below its December 2025 high and sits below both its 50-day and 200-day moving averages. This is not being traded as an equity. It is being traded as a long-dated call option on outcomes that have not happened yet.

Run the December call chain by strike and the positioning is extraordinary. Excluding a handful of dead adjusted-option strikes that carry open interest without live market – the $5 and $10 lines quote 0.00 bid against 0.01 ask with zero volume – 40.7% of Tesla's December call open interest sits at strikes of $600 or higher. That is more than 65% above spot, with four months to run.

The single largest call position on the entire board is the $990 strike, with 19,369 contracts. It is a genuine market, not a stale artefact: 51 cents bid against 56 cents offered, with 684 contracts traded on Friday. Its implied probability of finishing in the money is 0.4%. Behind it sit 12,782 contracts at $710 (1.6%) and 10,332 at $800 (0.9%).

Set that against the $500 strike used as this article's bull case – 17,062 contracts at a 10.5% probability – and the shape of the demand becomes clear. A meaningful share of Tesla's option flow is not expressing a view on whether the company earns more money next year. It is buying convexity on a discontinuous outcome: robotaxi at scale or Optimus at scale, with the risk that neither happens.

Having tracked how this stock reprices around news, that framing explains something the fundamentals cannot. Tesla fell 14.5% in a single session on 23 July, its worst day of the year, the day after publishing a shareholder deck that told investors the company was entering "its largest and most exciting period of investment." Alphabet fell 7% the same day. The selloff punished AI spending, traders said. Six sessions later Tesla printed its 52-week closing low of $298.32. It has since recovered 21.6%. Nothing about the earnings base changed across that 35% round trip.

These are not analyst targets. They are two of the most heavily owned strikes on Tesla's December expiry, and the probabilities are calculated from each strike's own implied volatility using N(d₂) – the risk-neutral probability of finishing in the money, which is a lower and more honest number than the option's delta that most commentary quotes instead.

A one-standard-deviation move at 45.1% implied volatility spans roughly $282 to $470 by expiry. Both the bull and bear cases sit just outside that band – which is precisely why both price near 10%, and why anyone quoting $500 as a base case is quoting a one-in-ten outcome.

The comparison that makes this concrete is a stock analysed two days ago. In the Apple (AAPL) profile bull and bear analysis, at-the-money implied volatility for the same expiry was 26.7% and the skew ran the conventional way: downside strikes dearer than at-the-money, upside strikes cheaper. Tesla inverts that. Same expiry, same market, opposite shape – because Apple's uncertainty is about a multiple and Tesla's is about whether two entirely new product categories exist at scale.

Tesla's own Q2 2026 shareholder deck is unusually direct about what it is doing with shareholder money. The company reported $0.4bn of GAAP operating income, $1.1bn of GAAP net income, and passed $100bn of trailing-twelve-month revenue for the first time. It also confirmed that Cybercab production began at Gigafactory Texas and that Tesla Semi remains on track for production this year in Nevada. Optimus construction started at Fremont after the Model S and X lines were decommissioned.

The deck identifies its own bottleneck plainly: battery pack capacity is "the main limiting factor to near-term vehicle production volume increase."

On the earnings call the same day, Elon Musk, Tesla's chief executive, was blunt about the hardest of the three ramps. "This is going to be the hardest product to scale manufacturing that we've ever made at Tesla, because everything on the robot is new," he said of Optimus.

Ashok Elluswamy, Tesla's vice-president of AI, gave the robotaxi programme's record on the same call: more than "380,000 miles of unsupervised Robotaxi across six cities in two different states with zero notable incidents."

The two figures are not the same measure. The shareholder deck states robotaxi "is now live in seven major metros," while Elluswamy's unsupervised-mileage figure covers six cities in two states. The deck's count includes supervised operation; the call's figure is the narrower unsupervised record. Anyone modelling the robotaxi ramp should use the second number, not the first.

Retail positioning has meanwhile been rotating. Retail investors piled into SpaceX while dumping Tesla, and the Musk complex has since seen 319 million SpaceX shares unlock. For a stock whose valuation rests on belief in a single operator's execution, where that operator's other assets are absorbing retail capital is not a trivial detail.

Here is the number that reframes the bear case entirely.

Tesla's trailing twelve-month diluted EPS is $1.08, derived from its SEC earnings-per-share filings with Q4 2025 backed out of the FY2025 annual figure. At $362.86 the stock trades on 336 times trailing earnings. At the $500 bull case it would be on 463 times. At the $250 bear case it would still be on 231 times.

A 31% decline in Tesla does not produce a cheap stock. It produces a stock at 231x earnings – still one of the most expensive large caps in the market. The bear case is not a de-rating to value; it is a partial deflation of an option premium. Investors reaching for $250 as a "floor" are reaching for a level that still embeds enormous expectations.

The synthesis that matters is on the earnings line, and it cuts against the bearish read. Tesla's diluted EPS collapsed from $4.30 in FY2023 to $2.04 in FY2024 to $1.08 in FY2025 – a 75% decline over two years. H1 2026 EPS of $0.45 is exactly flat against H1 2025's $0.45. The collapse has stopped. It has not reversed, and flat is not growth, the second derivative turned some time in the last twelve months while the narrative was still about decline.

Technically the stock has not confirmed that. At $362.86 Tesla sits just below its 50-day moving average of $365.86 and well below its 200-day of $403.32 – the structure of a downtrend that has bounced hard rather than an uptrend that has resumed. Realised volatility over the last 30 sessions is 57.6%, above the 45.1% the December options are charging, which is one reason those options do not look obviously expensive despite the headline level.

Friday's 5.14% move came from a regulator, not a factory. Nevada approved a permit allowing up to 5,000 Cybercab robotaxis in Las Vegas, alongside confirmation that Tesla Semi is heading to Europe.

That is the structural shift in this story. For most of Tesla's life the binding constraint was manufacturing. For the robotaxi business the binding constraint is jurisdictional approval, granted state by state and city by city, on timelines no company controls. A permit in Nevada does not generalise to California, and an incident anywhere resets the clock everywhere.

This is also why the AI-spending question is sharper for Tesla than for its peers. Tesla's investment programme runs through its relationship with xAI as well as its own silicon and factories, and the payoff depends on permissions that are not on any capex schedule. Compare that with a pure infrastructure build such as the NVIDIA profile analysis, where demand is contracted and the risk is digestion rather than authorisation.

Tesla carries an Alpha Score of 31/100, labelled Weak, at $362.86 in the Consumer Discretionary sector.

Realised volatility stays above implied into December. Tesla has printed six sessions of ±6% or worse in the last 90 trading days, including −14.5% and +8.5%. Thirty-day realised volatility is 57.6% against 45.1% implied for December. With robotaxi permits, Optimus milestones and Semi production all landing inside the window, the historical pattern says the December options are more likely to prove cheap than rich.

The stock resolves between $300 and $470 at expiry, with $350–$420 most likely. That band captures the bulk of the risk-neutral distribution: roughly 33% above $400, about 26% below $300, leaving the middle as the modal outcome. Both headline cases are one-in-ten events and should be treated as such.

The next repricing comes from a permit or an incident, not from earnings. Q3 results, which Tesla will post to its investor relations site, will show whether operating expenses decelerate, and that matters. Friday demonstrated the sensitivity clearly: a single state permit moved $70bn of market capitalisation. With 40.7% of call open interest sitting more than 65% out of the money, the positioning is built for exactly that kind of discontinuous headline – which is also why the downside is nastier than the 45% implied volatility suggests if the headlines run the other way.

The honest summary: Tesla is priced as an option, its option market knows it, and both tails are priced alike. The earnings base has stopped deteriorating, which the bears have not fully absorbed, and it has not started growing, which the bulls have not fully absorbed either. At 336 times earnings, the burden of proof sits with the ramps.