Switching out the heavy, predictable rotation of coal and gas turbines for the lean, fast-twitch response of batteries, wind, and demand-side flexibility, the European energy transition is often described as a shift in "muscles". But for this new muscle to work, the nervous system of the grid - the balancing markets, like PICASSO - must be equally fit.
Earlier this year at E-world 2026, we discussed the need to close the loop between physical assets and energy markets, a necessity that becomes undeniable as we move from a centralized past to a decentralized future. No market represents this challenge more vividly than PICASSO, the pan-European platform for the exchange of balancing energy from frequency restoration reserves with automatic activation (aFRR).
As we kick off a new series on cross-border balancing, we are starting with PICASSO - a platform that has delivered hundreds of millions of euros in efficiency gains while simultaneously making some traders feel like they are gambling blindfolded. On high-stress days, we have seen prices in Poland hit a staggering EUR 15,000/MWh - in this new reality of a coupled European grid, efficiency and volatility are two sides of the same coin.
But how does a system designed for continental optimization end up producing such extreme local outliers? Reconciling these numbers requires a look at the actual plumbing established by ENTSO-E and the participating TSOs. The primary role of PICASSO is to enable cross-border exchange and activation of standard aFRR bids through a common set of processes. By running a common optimization (AOF), it builds and clears a common merit order list for aFRR energy, considering cross-zonal exchanges as efficiently as possible. This process maximizes social welfare by pooling resources and reducing the total cost of keeping the grid at exactly 50Hz.
Through the Imbalance Netting Function (INF), PICASSO allows TSOs to avoid activating energy altogether if one region is long and another is short. It’s basically a continental game of "cancel the noise" before it requires physical - and expensive - intervention.
Since its launch in July 2022, a major accession wave has transformed the map of Europe. Initial joiners like Germany, Austria, and the Czech Republic formed the core, followed by the Netherlands in late 2024 and Italy’s significant re-connection in November 2025. The scale of this network is already staggering, with exchanged aFRR energy volumes hitting peaks of 60 GWh. By early 2026, over 26 TSOs are either operational or in the process of joining. But as the pool gets bigger, the waves are getting higher.
It’s a gamble
Despite its success in lowering average costs, PICASSO has become synonymous with extreme price volatility in certain regions. We have seen price spikes that look more like errors than market signals, leading some participants to describe the mechanism as creating unmanageable risk.
The most striking case occurred in Italy shortly after it joined in July 2023. The country observed extreme imbalance-price outcomes, with prices swinging to as low as -€9,000/MWh. This volatility was so severe that Terna was required to suspend PICASSO operations in March 2024 pending mitigation measures. This Italian shock sent a clear signal to the rest of Europe: integration without harmonization is a recipe for a blowout.
Similar stress tests appeared in Poland and the Baltic states following their accession in mid-2025. In September 2025, Estonia saw balancing prices crash to -€5,535.88/MWh, a drop that traders noted originated from bids in Poland. In Poland itself, the situation was compounded by a double-whammy in the local rules: the TSO (PSE) historically penalized renewable producers for imbalances even when those imbalances were caused by the TSO’s own curtailment of wind and solar output. When PICASSO's international price signals were layered on top of these local penalties, the risk for asset owners became exponential.
Traders are fed up. One famously compared the experience to a policeman getting into your car, telling you to speed, and then forcing you to pay the ticket.
This lack of transparency is particularly dangerous because PICASSO runs on a 4-second optimization cycle, yet market data for participants is often updated only once per hour. By the time a Balance Responsible Party (BRP) sees a price spike, the financial damage is already done, leaving them to trade blindfolded and manage their portfolio in a rear-view-mirror reality.
Is the algo broken?
When prices hit -€9,000, the immediate instinct is to blame the algorithm, but the math isn't the villain here. Economics professor Fulvio Fontii (quoted in Montel) and experts at Magnus Energy suggest the problem is the "growing pains" of fragmented national markets that only appear integrated on the surface.
As noted by Magnus Energy analysts Ksenia Tolstrup and Carlos Ubeda in a talk at E-world, the propagation effects of extreme price spikes have less to do with the algorithm and more to do with a lack of local liquidity, low liquidity at neighbors, and limited exchange capacity.
PICASSO is a marginal pricing system. If a small region becomes isolated due to the disappearance of real-time transmission capacity, the algorithm must clear local demand using whatever bids are available. If the only remaining bid in an isolated zone is an extreme outlier - such as a -€9,200/MWh bid - that outlier becomes the clearing price for the entire area.
The system currently acts as a compromise between different national systems rather than a true integration. Each market maintains distinct definitions and structures for frequency restoration reserves that can undermine cross-border trading. Until these markets adopt identical rules, the system will remain inherently unstable, characterized by price swings where countries become isolated during tight conditions despite theoretical integration.
Fixing the "muscle"
Fixing the turbulence in PICASSO requires a concerted effort from ENTSO-E, TSOs, and regulatory agencies to address the underlying fragmentation. Several levers are already being pulled:
1. During the suspension period in Italy, the aFRR implementation framework was updated at the EU level to include "elastic aFRR demand". This allows TSOs like Terna to better control responses to large price spikes, essentially giving them a safety valve when bid prices move into extreme territory.
2. The recently launched ALPACA platform represents an effort to harmonize aFRR capacity procurement across borders. It ensures that PICASSO has a deeper pool of bids to draw from. So while PICASSO handles the activation of energy (starting the cars), ALPACA manages the procurement of capacity (space on the highways). While it isn't a silver bullet for volatility, a more liquid capacity market reduces the likelihood that the algorithm reaches the empty end of the merit order list, where the most extreme prices hide.
3. True stability will only come when underlying local rules - such as balancing capacity procurement, Balance Service Provider (BSP) prequalification, and imbalance pricing - are fully harmonized. Some, like the Netherlands, have managed to avoid the worst of it by basing domestic imbalance prices solely on local activations, shielding their participants from foreign extremes. But that is only a temporary workaround.
What market participants can do
For BSPs and BRPs, this volatility is a double-edged sword. It represents a massive financial risk for those caught on the wrong side of a spike, but a significant opportunity for those who can move fast.
To handle this, participants must ditch manual workflows. The PICASSO optimization cycle runs every four seconds, while market data in some regions is updated only hourly. If your team is relying on slow data feeds, you are indeed trading blindfolded. High value is now placed on intraday optimization. You need to forecast and shift positions within minutes, not hours, to avoid the market's costly tails.
This is where advanced technical solutions like our EMoT platform come into play. By connecting physical constraints - like synchronization times, minimum runtimes, and ramp rates - directly to the market signal, operators can participate in these markets with confidence. Automation allows you to capture revenue in the mFRR, aFRR, and FCR markets without putting your hardware at risk from sudden, extreme instructions. We know these systems firsthand, as we supported the Baltic TSOs AST, Elering and Litgrid with integration of the Baltic Balancing & Capacity Market with PICASSO and are currently doing the same for Hungarian operator MAVIR.
Is volatility bad?
In the context of PICASSO, extreme spikes often signal inefficiency rather than true scarcity. But volatility is still the natural pulse of a renewable grid. However, more broadly, volatility is a natural byproduct of a grid powered by intermittent renewables.
We know some traders are annoyed, but it would not be fair to call PICASSO a failure. This platform shows us how far we still have to go in unifying Europe’s markets. It highlights exactly which markets are starved for liquidity or in need of structural reform. Most traders actually welcome a bit of movement provided they can predict it, or at least see it in time to move. The real friction lies in transparency.
As more participants join, we expect to see the catastrophic blowouts starting to settle, but the daily competitive grind is still running at a much higher resting temperature. For those with the right digital nervous system, that heat is an opportunity.
Ready to turn market volatility into a scalable business?
