Without addressing the issue, the result is a bad equilibrium where everyone is worse off.

To sum up:.- Slot timing should be equal to be fair. - In the current tokenomics, timing should be as small as possible. - While coordination is beneficial, it seems to be almost impossible to achieve. Continuity of choice (specific timing), large set of players make it worse.

What if model parameters would be different:. If Timing games played harder (more delay) → Effects are more disastrous. Gap between players and non-players is wider, which accelerates convergence. If issuance is lower → Break-even occurs at higher % of stake. To some extent,.

As finally the whole network is converted, everyone receives -1.4% less yield in absolute terms! . 7.1% vs 8.5%.

First entrants receive strictly better financial outcome, but the effect decays as more join. At 17% of players (in stake terms), there’s no difference from initial state (no players at all) and players go break even. But they cannot switch back: in this case their rewards rate.

Let’s consider a simple case, when validators decide to set slot duration at 400ms (non-player) or 500ms (player). On one hand, players receive higher block rewards, on the other, the inflationary reward is decreased. The figure below shows the net yield change for players and

They end up with an artificially increased share of block production over the epoch, measured in time. It’s as if a player effectively has a larger stake share than they actually do. Since the total value of transaction tips stays the same, timing games lead to the dilution of.

If everyone follows the same strategy, the tip-based portion of the yield remains unaffected. But what happens when someone diverges and starts playing timing games?.

Let’s consider a scenario where everyone adopts the same strategy, targeting a specific slot duration. The plots below illustrate how the average slot duration impacts the inflationary component (via the annual number of epochs), and how this, in turn, translates into changes in

Let’s do some napkin math to illustrate this. First of all, assumptions: .- Total Stake = 405M SOL.- Inflationary Yield = 7% .- Block Rewards Yield = 1.5%, tip flow is constant during the year.- Default slot duration = 400ms.

Early participants are incentivized by the desire to offer better terms to stakers or to gain higher rewards. Later entrants may be forced to follow suit just to stay competitive. Eventually, non-participants are diluted so much that they must start playing the same game just to.

So timing games directly impact epoch duration and can eventually lead to a net negative financial outcome for all validators. Why would validators or stakers engage in this behavior?.

Every epoch (432k slots), validators receive inflation-based rewards from the protocol. The longer the average slot duration, the longer the epoch lasts, meaning inflationary rewards are distributed less frequently.

While the protocol specifies a slot duration of 400ms, in practice, this duration is quite flexible - it can be extended up to the point where the block proposal is considered missed.

Simply put, staking rewards in Solana consist of two components: inflationary rewards and transaction tips earned through block proposals. Timing games emerge when a proposing validator artificially extends the block proposal process (slot time), thereby capturing more.

Timing Games in Solana: IBRL Prisoner's dilemma. Recently, the topic of timing games has sparked intense discussion within the Solana community. This phenomenon results in negative consequences for the network overall. While the operational downsides (e.g. instability in block

RT @ethStaker: The 2025 Staking Survey results are live!. We received 740 total responses from stakers of all types including solo stakers,….

Wrapping up the thread, I want to emphasize again that future Jito updates, and Solana more broadly, are both exciting and disruptive. It's amazing to witness how tech in different ecosystems converging to common meta. Looking forward to more details being unveiled! 🍿.

Nevertheless, malicious validators still can inject their own txs before execution. This vector is expected to be countered through black listing in case of sequence violation is spotted. Not elegant, but still viable. As a side note, this is not a problem in case of Ethereum PBS.

Will BAM solve toxic MEV? In many cases, yes. Malicious/incompetent RPC providers can still leak users transactions to sandwichers. But looks like BAM is capable to resolve attacks during the sequencing phase through plugins, or at least protection can work the same way like it.