LessonThis is a lesson restated in new words. Any figure in it is an example from the source, not this desk's record.
Fabio Valentini says that the Nasdaq-only model he has been building for six months is a way to take the intraday drift in short pieces, not a hunt for one large payoff. The host's introduction is specific: a professional scalper, alpha research for multiple seven-figure firms, and a podium finish in all four quarters of the futures audited challenge at the Robbins World Cup. Fabio has not properly tested the model elsewhere, and he refuses to say it transfers. He has other models that make money on the Nasdaq and lose it on the ES, and the reverse. Those introductions and results are his and the host's, not an audit, and he is explicit that his results are not a reason to assume yours.
Why it matters
The tool he calls Nasdaq effort is a simplification of which side is dominating. It studies imbalance, absorption, aggression, and the path of least resistance. If a given number of contracts moves price up more easily, the book is thinner on the upside and the market can explode there. Around that study he stacks value migration for the direction, delta and absorption for whether aggressive buyers or sellers are actually moving price, and large orders as a separate signature. A fully aligned filter, in the example he clicks through, is six trades, five of them target order, one a stop. An opening caption says five tick gain. The later, clearer count is five target orders and one stop. He names the overfitting risk directly. Stress a data set long enough and it will say what you wanted. The same tape still produces chop: a clean short, another clean short, a stopped long, then three stops in a row.
The idea in plain words
Friday, the session he dissects, was a consolidation day. His diagnosis of why people lose those days is the swing. They cannot tell when an impulse is finished, so they do not know when the second attempt is the buy. The open prints a first drive and a red cluster he already knew as strong absorption to the downside. Buyers pushed and were absorbed. The session then does what he calls the cage. Sellers were absorbed at the start, price came down, buyers were absorbed. The profile stayed balanced. He does not want the trade inside that balance. He wants a directional auction outside value.
The first setup he actually saw was at VWAP. For him that level is where reloading players put the maximum pressure to the downside. At the bottom, buyers print an aggression he can see. When price comes back into the value area there is another confluence: value-area low, and a line where buyers took complete control. Sellers tried to bring it back and failed. That failure is the absorption. Delta, he says, is the main force. A negative candle where buyers were aggressive means the buy power was absorbed. He also uses the delta candle to know when risk goes to zero. Enter, stop where he marked it, and when the other side takes control the stop comes to flat.
Nasdaq effort is the study he has kept for years, of when one side is actually dominating an area. Sellers taking control inside the box is his timing. A return to a place that already showed absorption, with a stack of two sell-control zones, is an extra confluence. On the buy he skipped the first push because it was too fast. The second one confirms at the close of the candle. Management is a trail under the last aggression. He expects the market to dip and resume. If it takes him out, he can reload when price is back inside the area. He is out below that low because the continuation he was positioned for failed.
Large prints are the third layer. A cluster of a few hundred orders might be one participant or many. Buyers pushing the auction with about three hundred contracts are a confirmation he uses to trail. One hundred and eleven contracts absorbed at the top is the kind of number he will show. A print around one hundred sixty-seven contracts, followed by buyers pushing the other side, is a pattern he associates with a bottom. Order flow is a data set. Buying the feed does not make someone profitable any more than buying a data set makes someone an analyst. The profile is the piece he calls objective in a way price action is not. Balance is what he calls the land of nowhere. Below it is the discount.
How to use it
He describes the hold as scalp to intraday. Short-term positions are how he takes the drift. Aggressive scalping is a different job. The tool he calls the order track answers, candle by candle, who is dominating and where the value area is migrating. A downward drift has value shifting lower. When that value starts switching higher, and the candle that was seller-dominated gives way to buyers, that is where he loads a one-to-one. The reason he prefers several one-to-ones to a single one-to-five is the regime. If the day's direction is understood, every movement in that direction can be scalped. Three or four one-to-one positions can pay inside a market that is also consolidating. A one-to-five can park the trader in one position for a long time and then do nothing. He is honest that the next one-to-one can stop out before the idea is wrong in any larger sense.
The sell is the same machine in the other direction. Value switching lower, sellers dominating, stop above the value area he is protecting, one-to-one. A sell candle that closes back inside is not yet the next short. A break below is. A green candle while value is trying to switch lower is a skip. If the point of control has stepped down from the previous candle, value is migrating lower. Absorption of a failed break can justify pressing a one-to-two. The objective of the model is not the large payoff. It is a consistent equity curve, and only then, maybe, risking the gain already made on a larger position.
The win rate is not a single number. On a drift he can print three or four targets in a row, leave one unfilled, flip because value switched, and print more targets on the way down. He calls the win rate of the model really high, and he will not automate it. An automated version leaves the risk where the entry candle put it. He sees a large buyer absorption and trails the stop to zero. That refinement is the discretion. On a real drift day that can be three or four rolling positions off the initial risk of one. There are also negative days. In extreme chop the win rate falls to around forty to fifty percent, which at one-to-one, plus commission, is break-even. He does not call it a holy grail.
Value area is the addition of the last six months, and he thinks it is the part that tells him how sustained the drift is. A big order can be scattered through several candles, so the next candle retraces and a trader who chased the print is flat. Asked what matters most, he says value area together with delta, and the last tweak is the Nasdaq effort area itself. Validation was at least three years of data, then three months out of sample, watching, because the trade count is high. Three months might collect about three hundred trades. He did not publish it as a standalone strategy. It is a proprietary study and a set of indicators, meant to sit next to a model the trader already has. He disagrees with relying on one model. He runs a longer-term mean-reverting trend approach beside this one. Some days this model is dead and the longer one works. Some days both are dead and a crypto model works. Across a lot of testing he has not found a single model that is profitable in every environment.
What usually goes wrong
Consolidation is the weakness he accepts for every trend-following system, this one included. In a consolidation week the goal is to protect capital, not to blame himself. Commission is a known cost, and the impact is larger on short-term trades. Partials are the piece he and the host both killed for this model. The trail is two Nasdaq-effort boxes, not a scale-out. He is in at the first green box. The second box prints and the stop goes to breakeven. Another box, and the stop steps under the second. He tested partials. On Nasdaq they remove him from an asset that can trend all day. A partial can flat him, and then two hundred points of rally are gone. On currencies he does take partials, because that market mean-reverts more. Nasdaq, in his description, has an upward directional skew. The host reports the same result from the systematic side: partials cut the gains by more than they cut the risk. Someone who makes the indicator the authority and stops deciding the direction is the error he is guarding against. If the filter could delete every consolidation, he says, it would not be a system.
Questions people ask
Why several one-to-one trades instead of one large target?
If the drift is understood, each push in that direction can be a one-to-one. Three or four of them can pay on a day that is also consolidating. A one-to-five can sit in one position and then do nothing. In extreme chop he says the win rate falls to around forty to fifty percent, which at one-to-one plus commission is break-even. "Really high" is his phrase for the drift, not a published percentage.
Does he say the model works on the ES?
No. It is Nasdaq only. He has not properly tested it elsewhere and refuses to say it transfers. Other models of his make money on one of those indexes and lose it on the other. Correlation, he says, is not the same microstructure.
What did the out-of-sample watch involve?
At least three years of data, then three months out of sample. Three months might collect about three hundred trades. He did not publish a standalone strategy. The study is meant to sit next to a model the trader already has, and he keeps a longer mean-reverting approach, and a crypto model, because he has not found one model that is profitable in every environment.