CoinSpiderMethodology 1
Value Depressions: How CoinSpider Finds Overlooked Small-Cap Coins
A deep drawdown is not the same as cheap. This paper describes the three-layer framework CoinSpider uses to find what we call value depressions: first confirm a coin is tradable, then measure where its price sits, then check for catalysts and risks. We only study small-cap Binance spot coins that pass all three layers, and we explicitly exclude the rest — above all, coins that are still sliding, which we call abandoned rather than overlooked.
Key points- A value depression is a coin at a long-term price low that still trades for real and could plausibly be rediscovered
- Liquidity comes first: value only counts if you can get in and get out
- We look for overlooked coins, not abandoned ones
The problem: small caps need their own method
Binance lists hundreds of spot pairs, and most small caps go months without anyone talking about them. Chasing whatever is trending barely works here: by the time a coin is trending, it is rarely still at the bottom. The losers list is no better — many coins fall because the project has stalled, tokens keep unlocking, or liquidity has dried up.
So we need a framework that can be run repeatedly and explained, and that answers one specific question: right now, which coins sit at a long-term price low, still trade for real, and could plausibly attract attention again? We call coins that meet all three conditions value depressions.
Layer 1: tradability
Before talking about value, a coin has to be tradable. CoinSpider first applies hard filters that remove coins carrying Binance risk tags such as Monitoring or Seed; coins listed too recently to judge; coins with persistently thin trading, sudden volume cliffs or almost no turnover; coins whose volume is wildly out of line with their market cap, a sign of wash trading; and coins that have just rallied hard.
The logic is simple. Cheap with no liquidity is only cheap on paper: slippage eats the discount on the way in, and there may be no buyer on the way out. We would rather miss these than hold them. The exact thresholds are strategy parameters and are not published, but every scan applies them to every candidate in exactly the same way.
Layer 2: price position
Coins that pass the first layer receive a potential score. Four dimensions describe price position: valuation (where the circulating market cap sits within our research range), drawdown (how far it has fallen from its peak on Binance daily candles), base pattern (whether it has stopped falling and started to build a base) and volume (whether trading is steady and beginning to pick up).
Drawdown is the most misunderstood. A coin down 95% can easily fall another 95%. So we measure "fallen a long way" separately from "has stopped falling": the first is a precondition for opportunity, the second decides whether now is a reasonable entry. Figure 2 contrasts two typical shapes. One is still in a downtrend, every bounce making a lower high — an abandoned coin. The other has a narrowing range, lows that no longer move down and trading that is starting to recover — an overlooked coin. We study the second.
Layer 3: catalysts and risk
A low price says a coin is cheap, not that it will be repriced. The third layer asks three questions. Is the project's sector one that money is returning to (narrative attention)? Will circulating supply and upcoming unlocks keep weighing on the price (tokenomics)? Are there on-chain signs of accumulation, such as large amounts leaving exchanges (on-chain flows)? On top of that, if other major exchanges show a premium over Binance, a volume surge or a new listing first, buying elsewhere has moved ahead, and the coin earns extra points; a discount in the same window cancels them.
None of these is a certain signal. A narrative can take a long time to arrive, and an outflow may just be an exchange moving its own funds. That is why the top-scoring coins then go to the AI review council, where four reviewers with different roles judge each one against the evidence, and the risk officer can veto on red flags such as delisting, unlocks, a stalled project or fraud.
Why it is designed this way
The core idea is to rule out first and choose second. With small caps, avoiding one fatal mistake usually matters more than catching one opportunity. The layered structure means every excluded coin has a stated reason and every coin sent to review has a traceable score breakdown — tap any coin on the potential board in the app and you can see those components and the key data behind them.
The framework is also deliberately narrow: Binance spot small caps only, no futures, no short-term trend chasing. Staying focused lets us accumulate data under one consistent definition and test every day whether the method works. That is what the daily research log is for.
Limitations
A screening framework can only improve the odds; it cannot remove risk. A value depression can stay depressed for a long time, and data sources have gaps and delays. Every day the research log publishes real scan, review and live-trading data, failures included, so you can judge for yourself whether this approach suits you.
