Mathematical Money | October 12, 2026
On 1 October I posted the levels I was watching on $COIN$ , four zones drawn off nine months of price action, with no model behind them beyond where the stock had actually spent its time.
In the nine days since, four separate houses published new price targets on the same company. Every one of them was a raise. Barclays went to $149, Bank of America to $203, Goldman Sachs to $244, Morgan Stanley to $258.
Then I went back and put their numbers on my zones, which I had not expected to be interesting and turned out to be the most interesting thing I have looked at this week.
Where their targets land on my map
Barclays at $149 sits inside what I called the liquidity grab zone, $145 to $160 , the area I described as volatile and matching the previous lows.
Bank of America at $203 sits right at the foot of what I called the waiting to shoot zone, $200 to $220.
Goldman at $244 and Morgan Stanley at $258 are off the top of my map entirely. I did not draw a zone up there because the stock has not been there in nine months.
So the four of them, between them, span from my bottom zone to above my top one. After all four revised upward.
That is worth sitting with. The spread between the highest and lowest target on the street is $109, on a stock trading at $179.39. One house sees 17% downside and another sees 44% upside, and they are reading the same filings.
The zones held up
I drew those levels off the chart rather than off a spreadsheet, so afterwards I went and counted. Across the 188 sessions from 2 January to 1 October, the $160 to $180 zone took 63 days, a third of the entire period, comfortably the most-visited of the four. That is what I had called the primary support area and it is what the data says it was.
Below it, the $145 to $160 liquidity grab zone took 31 days. Above it, the $180 to $200 middle took 56, and $200 to $220 took 22. The stock closed above $220 on fourteen days all year and below $145 on two.
Since that post it has gone from $189.29 to $179.39, down 5.2%, which carried it out of the middle zone and down to the boundary of primary support. That is the kind of move the framework describes, and none of the predictive credit belongs to me. I said the middle was directionless, and it was.
Why they can't agree, and why that is the useful part
Coinbase is three businesses wearing one ticker, and which one you think dominates decides your number.
There is the exchange, earning transaction fees that swing with volumes and therefore with crypto prices. September volumes surged. Early October already points to a pullback. If you model Coinbase as an exchange you are really forecasting bitcoin, and you should be honest that you cannot.
There is interest income on USDC balances, which is steady, contractual, and indifferent to trading volumes. It rises when rates rise, which in this environment is a feature.
And there is payments, custody and institutional services , recurring, far less cyclical, still small. The bull case is that this becomes the main thing and the market stops pricing the company as a leveraged bet on crypto.
A $149 target is someone modelling an exchange in a cooling market. A $258 target is someone modelling infrastructure early in a transition. Neither is stupid.
The plan did not change. The strikes walked up.
On 1 October I said I would keep selling puts below $180 and running staggered covered calls between $175 and $190, and that I did not expect a bull run before the third-quarter numbers. The puts are unchanged, short at $177.50 and $180, with the stock at $179.39.
The calls have moved, deliberately, and this is the part worth explaining.
On 21 September I sold calls at the $190 strike for $24.62. By 8 October the stock had come down and those were worth $3.25, so I closed them and wrote new ones at $202.50 for 6 November, taking a credit on the switch.
That is a ladder, and it is the thing I am actually doing rather than sitting in a fixed band. Each time a call decays I do not rewrite at the same level. I walk the strike up. $190 became $202.50, which is a ceiling twelve and a half dollars higher than the one I had before, and I was paid to move it.
The trade-off is explicit and it runs the other way on premium. The $190 calls brought in $24.62. The $202.50s brought in $4.30, barely a sixth as much. I am accepting far less income per contract in exchange for far more room above me, because what I want from this position between now and the earnings date is not maximum premium. It is to still own the shares if something lifts the whole sector.
Which does mean $202.50 sits outside the $175 to $190 band I named nine days ago. It sits at the floor of the zone I called waiting to shoot. That is not me abandoning the plan, it is me raising the ceiling as the position allows, and I would rather describe it that way than pretend the band was a rule.
The honest limit of a ladder: it only works if you keep climbing it. If Coinbase runs hard past $202.50 before 6 November, I am capped there, and the only answer is to roll up again and pay for the privilege. Each rung costs something when the stock is moving. I think that is the right trade into a catalyst I cannot time, but it is a trade, not a free option.
The bit that looks like a rule being broken
Something above should have bothered you. My cost on the shares is $185.05, and I just said I wrote calls at $180 , below what I paid.
I have a standing rule against that, because assignment under your cost locks in a loss on the stock. So either I broke my own rule, or the rule is more subtle than it reads.
It is the second. Those $180 calls were sold for $13.59. If the stock finishes above the strike and the shares go, I receive $180 plus the $13.59 already collected, an effective sale at $193.59 against a cost of $185.05. That is a profit, despite a strike nine percent below where I bought.
The strike is not the sale price. The strike plus the premium is. On a quiet stock the premium is far too thin to bridge a gap like that and the rule applies exactly as written. Sixty-day volatility here runs at 77%, and at that level the premium does the bridging. Refusing to write below cost would mean leaving the best-paid strikes alone for no reason beyond a number in a spreadsheet.
Which is why I keep it as a question rather than an instruction: is the effective sale price above my cost? Here it is, comfortably. When it is not, I do not write the call.
What I'm watching
Earnings on 28 or 29 October, and the first line I will read is not revenue. It is the mix , how much came from transaction fees against subscriptions and services.
If the non-trading share keeps growing, the $258 case strengthens and this stops being a bitcoin proxy. If the quarter is carried by September's volume spike, Barclays is closer to right and it is an exchange that had a good month. That single disclosure is most of the $109 gap between the top and bottom of the street.
Two things I'd like other views on.
When targets on one company span from well below the price to well above it, do you read that as information or noise? I have started treating wide dispersion as a signal in itself, usually that the market has not settled what kind of business it is looking at, which is a better environment for selling premium than for picking a direction.
And on laddering: when you roll a covered call up, do you take the credit or pay to move further? I have been taking credits and climbing slowly. The alternative is a debit roll that buys more room in one go, and into an earnings date I am genuinely unsure which is right.
Price targets as reported 5–9 October 2026. Zone day-counts and prices from Tiger daily bars, 2 January to 1 October 2026.
#crypto #options #wheelstrategy #coveredcalls
Stop guessing. Start calculating.
Live to fight another day. 🤙
Comments