Institutional Trading Playbook

The useful way to manage this is to stop treating impact as a single cost number and start treating it as a dynamic response. The Impact Triangle Model does exactly that: it links three forces that move together during execution - market impact (how your trades move prices), liquidity (how much trading capacity exists at each price level), and timing (when you trade relative to evolving market conditions). In a healthy market, these forces balance. In stressed markets, timing becomes dominant, and liquidity can “thin out” faster than your internal assumptions.

Here’s the concrete problem this solves. Suppose you want to buy 1,000,000 shares over the next two hours. Your blotter says you can average 200,000 shares per 20 minutes. Your risk desk says the stock trades about that volume. Yet when you actually run the order, the realized price spreads widen, and your fills come back worse in a way that your simple participation model doesn’t predict. You need a framework that tells you how to adjust your pace when liquidity changes and when your own trading starts to move the tape.

Ava Chen, 34, equity execution trader at a buy-side firm, runs into this every time markets shift from “normal” to “busy.” She doesn’t blame one input. She checks whether the market can absorb the order at the current spread and depth, whether her pace increases the pressure, and whether the clock matters because volatility and liquidity change minute by minute. The Impact Triangle Model gives her a language for that check that doesn’t rely on vague “market feels heavy” opinions.

How It Works: The Impact Triangle Model You Can Run During Execution The Impact Triangle Model treats your execution as a balancing act between three vertices:

1) Market impact: the price concession (or improvement) created by your trading and by the market’s reaction to your displayed size. 2) Liquidity: the available trading capacity near the market (depth, spread, and how quickly orders replenish). 3) Timing: the schedule you choose - how fast you trade, when you trade relative to events, and how you respond to changing volatility.

You can’t control impact directly, but you can control your interaction with liquidity and timing. When liquidity tightens (spread widens, depth thins), the same order pace creates more impact. When volatility rises, liquidity often becomes more fragile, so impact grows even if your pace stays constant. That’s why execution plans that worked yesterday can degrade today.

Use the model with a simple loop: measure the triangle in real time, adjust your pace, and keep your order’s signaling aligned with current liquidity.

1. Quantify liquidity at the moment you send size. Track spread and visible depth (order-book depth at or near the touch) and confirm how quickly the market replenishes after trades. If the spread widens by 50% and depth drops materially, you treat liquidity as lower even if total tape volume still looks “fine.”

2. Estimate marginal impact for your current pace, not your original assumption. Compare your expected execution price (from your internal model) to the first window of realized fills. If your slippage per share worsens after you increase pace, you learned that impact rises with your interaction - so you stop assuming linear costs.

3. Set a timing schedule that matches the liquidity regime. Trade faster when liquidity looks stable and slower when it thins, but also avoid “stalling” during adverse drift. Use short decision windows (for example, 10-15 minutes) so timing adjustments happen before the order drifts too far.

4. Close the loop: adjust pace based on how the triangle moves. Increase pace only when liquidity holds and observed impact stays contained. Decrease pace when spread/depth deteriorate or when your observed slippage escalates beyond tolerance.

Ava’s day-to-day example: she runs a buy program that started with a participation target, but the first 15-minute window shows widening spread and slower replenishment. She doesn’t just lower participation blindly. She checks whether the slippage per share increases as she trades and whether the order book stops “repairing” after her prints. When both happen, she reduces pace and shortens her aggressiveness, so she trades less aggressively into thin liquidity while still maintaining a path toward completion.

To make this operational, you need a pace-adjustment rule that ties directly to what you can observe. Here’s a concrete way to do it: pick a tolerance for realized slippage versus your expected range during each 10-15 minute control window. If realized slippage exceeds tolerance while spread widens and depth thins, you reduce pace. If realized slippage stays inside tolerance and liquidity looks stable, you keep pace or slightly increase it. This keeps timing and liquidity adjustments tied to observed impact rather than gut feel.

Putting It Into Practice: A Real Execution Walkthrough With Expected Outcomes Below is a practical walkthrough for a single large order. I’m going to keep the numbers explicit so you can map the logic to your own systems.

Scenario: Ava Chen needs to buy 1,000,000 shares of a liquid equity. She has 2 hours to complete the order and she wants to manage execution quality against both market drift and self-induced impact. She plans to use a decision window of 10 minutes and checks spread/depth and realized slippage in each window.

Step-by-step execution loop 1. Pre-trade: set an initial pace that matches “normal” liquidity. Assume you expect to complete 1,000,000 shares in 120 minutes. That implies an average of 8,333 shares per minute. You don’t send all at once; you send a steady schedule that you can throttle quickly.

Expected outcome: you start with a pace your model can handle when liquidity behaves normally.

2. Start execution and measure the first window (minutes 0-10). Record: - current best bid/ask spread - visible depth near the touch - realized execution price versus expected - whether fills consumed the visible levels and whether the book replenished quickly

Expected outcome: you calibrate your marginal impact for the current regime.

3. Apply the Impact Triangle decision rule at minute 10. - If spread widened and depth thinned and your slippage per share exceeded tolerance, cut pace (reduce shares-per-minute by a meaningful step, not a token tweak - think “large enough to change the interaction”). - If spread and depth held and slippage stayed inside tolerance, keep pace steady. - If slippage worsened but spread/depth stayed stable, reduce aggressiveness (you likely pushed the market via order placement/signal rather than liquidity shortage).

Expected outcome: you stop paying increasing marginal cost caused by the triangle shifting against you.

4. Repeat the loop every 10 minutes until completion. In each window, update two things: - liquidity score: spread and depth trend - impact score: realized slippage trend versus tolerance

Expected outcome: your completion time remains on track, but you avoid runaway impact when liquidity thins.

5. Handle the final 15 minutes with a completion bias rule. As you approach the end, you switch from “pace chasing” to “completion control.” If you lag target completion by more than your allowed threshold, you increase pace gradually while monitoring slippage and spread/depth. If you lead completion, you slow down to avoid overshooting into thin liquidity.

Expected outcome: you finish the order without turning the last-minute scramble into a large impact event.

Quick checklist - Confirm spread and visible depth at the touch before you send the first slice. - Log realized slippage versus expected in each 10-minute window. - Treat widening spread + thinning depth as “liquidity down,” even if total volume looks okay. - Cut pace when both marginal impact and liquidity deterioration move together. - Slow aggressiveness when slippage rises but liquidity metrics stay stable. - Use a separate completion bias rule in the last 15 minutes.

A key differentiator in the Impact Triangle approach is that you don’t adjust based on one metric. You adjust when the interaction tightens: liquidity deteriorates and your marginal impact worsens at the same time. That combination usually drives the “why did we get worse?” events.

What to Watch For: Mistakes and Edge Cases That Break the Triangle Even strong execution teams break the triangle when they treat it like a static diagram. Here are the most common failure modes and how to fix them quickly.

Mistake: You chase participation while liquidity collapses Do this: Reduce pace immediately when spread widens and depth thins, even if your participation target says you should keep going. Your participation math assumes the market can absorb the flow at the prices you plan to hit. When liquidity thins, the market cannot absorb the same flow without moving price. Not this: Keep pace because “we’re on schedule.” You’ll often discover that the schedule itself becomes the problem: your order keeps consuming the same shrinking liquidity and you pay for it repeatedly.

Mistake: You update impact assumptions too late Do this: Calibrate marginal impact using the first 10-minute window and then every 10-15 minutes. If you wait for end-of-day results, you miss the regime change and you lock in the damage before you respond. Not this: Keep using the pre-trade impact estimate because it looked reasonable at the open. The triangle shifts with volatility and order-book repair speed, and those changes show up early.

Mistake: You treat timing as “set and forget” Do this: Use a completion bias rule for the last 15 minutes that ties pace changes to observed slippage and liquidity, not just remaining shares. If you lag, increase pace gradually while monitoring spread/depth; if you lead, slow down to avoid paying a late-stage liquidity premium. Not this: Slam the remaining shares with a single push near the close. That creates a timing spike that often overwhelms replenishment and turns marginal impact into a step function.

If you want a simple mental test, run this after each window: “Did the market absorb my flow with stable cost, or did my flow start to change the cost structure?” When the answer becomes “my flow changed it,” you treat timing and liquidity together and you adjust pace and aggressiveness as one decision.

The takeaway is straightforward: market impact doesn’t live alone, liquidity doesn’t stay constant, and timing isn’t just a calendar choice. The Impact Triangle Model forces you to watch the interaction in real time and respond before the cost curve breaks. As you build this habit window by window, you’ll find your execution plans stop drifting and start behaving like systems - repeatable, monitorable, and controllable.

End of chapter one. 7 more chapters in the full book.