Ramping Capital Into a New Strategy

A brand-new strategy rarely goes from zero to its full target size on day one. Instead it gets a ramp schedule: a pre-agreed sequence of capital increases, each one conditional on the strategy clearing a specific bar of live performance and operational stability. The logic is simple — a backtest and even a few weeks of small live trading can look promising and still be wrong in ways that only show up once real money and real market impact are involved. A ramp schedule turns "it looks good so far, let's give it more" from a discretionary, emotion-prone decision made in the moment into a structure that was agreed on before anyone knew how the strategy would actually perform.

A typical ramp might start a strategy at 10% of its eventual target allocation, hold there for four to six weeks, and only step up to 25%, then 50%, then 100% if each stage clears agreed checks: live P&L within a plausible range of what the backtest implied, no unexplained divergence between simulated and live fills, no risk-limit breaches, and no operational incidents. Each step is a gate, not a formality — failing a gate means pausing at the current size, extending the observation window, or in some cases reversing back down a stage rather than proceeding.

The reason this matters is that a strategy's true costs and true capacity are things you can only observe by trading it, and both tend to bite harder as size increases. A strategy that looks fine at a small clip can show much worse slippage or crowding effects once it's moved up to a size where its own orders start to matter — ramping in stages catches that before the firm has committed its full intended capital to something that doesn't scale as assumed. It also limits the damage of the far more mundane risk that new strategies carry: bugs. A sizing error, a mis-mapped instrument, or a mistimed signal is much cheaper to discover at 10% of target size than at 100%.

Ramp schedules are usually set by the same body that approves the strategy in the first place, often a capital allocation committee, and are written down in advance rather than negotiated stage by stage. This matters because a strategy that is performing well creates pressure to accelerate the ramp, and a strategy that is performing poorly creates pressure to slow it down or reverse it — both are exactly the moments when having a rule fixed in advance protects against overriding good judgment with recency bias.

What this means in practice

The specific milestones vary by firm and by strategy type — a slower-turnover fundamental strategy might ramp over months, a high-frequency strategy over days — but the underlying idea is constant: separate the decision "should this strategy exist at full size eventually" (made once, up front) from the decision "does live evidence support increasing size now" (checked repeatedly, against pre-set criteria). Strategies that fail a ramp gate are not necessarily bad strategies; sometimes the fix is a smaller target size, a different execution approach, or simply more time at the current stage before trying again.