Pre-dispatch prices show how the market expects prices to clear before real-time dispatch. For batteries, the forecast data shows what operators saw when they made bidding decisions, not just what settled afterwards.

Operators use pre-dispatch to position state of charge, update bids and respond to expected price events. Looking across successive forecast runs shows how those signals formed, strengthened or disappeared before dispatch.

You can now query Ko for the full 30-minute and 5-minute pre-dispatch history from November 2020, alongside settled dispatch prices and bid stack data. That makes it possible to compare forecasts, outcomes and bidding strategy in one place.

Pre-dispatch is AEMO’s forward price solve

Pre-dispatch is the NEM’s live forward price signal. It is the actual NEMDE solve for future intervals, using the bid stack and system conditions available at the time. It is calculated in the same way as the dispatch price, but for intervals that have not happened yet.

Battery operators will often use their own internal price forecasts as well, but pre-dispatch remains the common market benchmark. It shows the forward price signal available to every participant at the same point in time. Retailers and gentailers also use it to manage short-term hedge exposure.

Operators use that signal to help decide when to charge, discharge or preserve charge. The signal is strongest when the forecast is close to dispatch and the system is stable. Longer lead times carry more uncertainty, especially when rooftop solar, wind output, demand or interconnector flows are moving quickly.

​Ko can query pre-dispatch history from November 2020

• 30-minute pre-dispatch forecasts: AEMO’s half-hourly regional energy and FCAS price forecasts, re-published every 30 minutes out to roughly 40 hours ahead. All five NEM regions, November 2020 to present.
• 5-minute pre-dispatch forecasts: the near-term forecast, re-published every 5 minutes out to about an hour ahead. Same regional coverage and period.
• Settled dispatch prices: the actual 5-minute spot prices each region cleared at, used to compare forecast prices with realised outcomes.
• Bid stack data: generator price and volume bids for each 5-minute interval, used to compare bidding strategy with the pre-dispatch signal available at the time.

​How to analyse NEM pre-dispatch data with Ko

​The examples below were produced by asking Ko questions against the NEM pre-dispatch price tables. Ko generated the SQL, queried the data, and produced the written interpretation. All charts were built from the same underlying data.

Each section shows the prompt given to Ko, Ko’s written output, and a chart built from the data Ko retrieved. You can ask Ko the same questions directly, with results reflecting the latest available data.

​Output: Each line shows a pre-dispatch run for New South Wales on 25 May 2025, with the settled RRP overlaid.

Negative midday prices were visible early. The midnight run was already forecasting negative prices from around 10:30 to 12:30, broadly in line with the final outcome.

The trough was shorter than early forecasts suggested. Early runs expected negative prices to last until around 13:30, but settled prices recovered earlier.

The morning peak was revised up closer to dispatch. Early runs forecast $65-75/MWh, whilst settled prices reached $72-77/MWh.

​​Output: Victoria’s 7:30 pm interval on 3 February 2025 was forecast near the market price cap for most of the day, before the signal was revised down close to dispatch.

Pre-dispatch held near the cap through the day. Runs from 7:30 am to 4:30 pm forecast the 7:30 pm interval near $17,500/MWh, with a mid-morning dip to around $13,300/MWh.

The forecast reset around 5:00 pm. At roughly 150 minutes out, the forecast fell from around $12,600/MWh to $565/MWh.

Later runs stayed lower. Subsequent runs through to the final 6:00 pm pre-dispatch stayed around $313-370/MWh, suggesting the expected scarcity had largely cleared before dispatch.

​Follow-up: The 30-minute settled price for the interval was $6,034/MWh, driven by 5-minute dispatch prices above $11,000 in the opening minutes. Pre-dispatch forecasts a single price for the full trading interval and cannot resolve that kind of intra-interval volatility.

Output: The four-hour-ahead pre-dispatch gap in South Australia’s midday window has moved through three regimes since 2021.

2021: under-prediction. The forecast expected more negative midday prices than actually cleared, with the gap peaking at +$81/MWh in Q3 2021.

2022: over-prediction. The sign flipped during the energy crisis. Q2 2022 recorded the largest gap at around $860/MWh, with the four-hour-ahead forecast averaging $1,001/MWh against settlement of $138/MWh.

2023-2026: structural over-prediction moderated. Forecasts remained higher than settlement during midday, but the gap narrowed from -$149/MWh in Q3 2023 to between -$13/MWh and +$8/MWh in 2026.

The trend toward zero suggests pre-dispatch has become better calibrated to South Australia’s midday solar suppression, or that solar’s incremental impact on midday prices has plateaued

​Output: Wallgrove’s bidding on 14 May 2025 followed pre-dispatch signals, with discharge capacity offered only when forecast prices rose.

Overnight bids stayed withdrawn. From midnight to 6:15 am, Wallgrove’s cheapest discharge band sat near the market price cap, whilst New South Wales pre-dispatch prices stayed around $87-110/MWh.

Morning bids moved into the ramp. From 6:20 am, Wallgrove lowered its cheapest band as pre-dispatch prices climbed toward the morning peak, then withdrew again after prices receded.

Evening bids stepped down ahead of the ramp. From 3:55 pm, Wallgrove progressively lowered its cheapest bid from $450/MWh to $35/MWh as pre-dispatch prices rose, before withdrawing again after 6:25 pm.

The pattern points to a price-targeting bidding strategy, with capacity made available when pre-dispatch signalled higher prices.

​Analyse NEM pre-dispatch data with Ko

• How did the pre-dispatch forecast for Queensland's evening peak on [date] evolve over the six hours before dispatch?
• Which days in New South Wales saw the biggest pre-dispatch revisions between the 6-hour ahead and 30-minute ahead forecast this year?
• How has the gap between Victoria's pre-dispatch forecast and the settled price changed quarter by quarter since 2021?
• For [a NEM battery], chart how its bidding lined up against pre-dispatch prices on a specific day.
• How did the 5-minute pre-dispatch forecast for [a region] converge in the hour before a price spike?