James 'SplitSuit' Sweeney is a poker player, coach, and author dedicated to helping players think deeper and win more.
As a co-founder of Red Chip Poker and creator of numerous strategy resources — including The Poker Bank YouTube channel — he’s known for turning complex poker theory into clear, actionable advice.
In the new book, The Exploitative Edge, we discuss the role GTO solvers play in crafting exploitative lines. Many players who attempt to use solvers to accomplish this end up using the “node locking” feature.
While this seems solid on the surface, there are some issues.
The common misconception is that a single node lock can be used to generate a maximally exploitative strategy. Node locking is a process by which we force the solver to use a particular strategy at a specific decision node.
For example, a given population folds 42% of the time when facing a 33%-pot c-bet on an Ace-high or King-high texture. We could place a node lock on the game tree which forces the BB to fold exactly 42% of the time rather than the GTO-approved frequency of 34.9%.
At this stage, we can rerun the solve. Every aspect of the solving algorithm remains the same, aside from the higher flop folding frequency due to the node lock.
Nobody's perfect
The question is: what does the solver give back to us?
A common belief is that the solver output now represents the maximum exploit against the BB overfolder. Put simply, it does not. A better description might be somewhat exploitative.
The solver, by its very design, assumes that the BB proceeds by playing perfect GTO poker at every other decision node.
This causes two significant problems with the output:
- The solver can exploit the BB only while remaining within the confines of being immune to counter-exploitation at every other decision node.
- It is not realistic to assume that a player overfolding the flop is not also making mistakes at other decision nodes.
If we wanted to use the solver to generate a true maximum exploit, we would need to lock every node of the decision tree with villain’s exact strategy (assuming we even have access to such data). While such an approach is theoretically possible, it is much more complex than even advanced poker players realize.
Zoom out
If you ever find yourself attempting this, you will quickly realize that this is not what a GTO solver is designed for.
This is not to say that a single node lock as part of a GTO solve is not useful. It can give us insights into how to adjust against a pool/villain imbalance. But it has never been designed as a tool for generating maximally-exploitative strategies.
So the next time you find yourself getting bogged down in node locked solver exploration, attempting to find perfect exploits…zoom out and be sure to think about your opponent’s strategy at all other nodes as well.
Likely, your time is better spent either a.) using the solver as-intended or b.) analyzing player pool data to find exploits in the stats. If you’re interested in the latter option, be sure to pick up your copy of The Exploitative Edge to learn how 👍
Follow James on X @SplitSuit and visit splitsuit.com for more poker strategy articles, books, and courses.
