Botmaster Phil Godzin’s Sudan conflict deaths forecast, Oct. 3, 2025

How many state-based conflict deaths in Sudan will be reported by ACLED in 2025?

Phillip Godzin’s pgodzinai predicts today:

Less than 1,000: 0%
Between 1,000 and 3,000: 0%
Between 3,000 and 5,000: 0%
Between 5,000 and 8,000: 1%
Between 8,000 and 12,000: 2%
More than 12,000: 97%

10–13 minutes read (This is a condensed version of pgodzinai’s results. Full results here —>

Perplexity response:

# Current Conflict Status
The Sudanese civil war between the SAF and RSF has continued with **high intensity throughout 2025**. The siege of El Fasher represents one of the most active battlefronts, with documented major engagements in January, March, and April 2025. The January 24 assault alone reportedly resulted in over 400 RSF casualties, with the SAF claiming to have “thwarted over 170 attacks on the city, killing over 15,000 fighters”.

The conflict has expanded geographically, with fighting shifting toward Nyala in South Darfu, and recent tribal alignments potentially escalating ethnic violence. Arab tribal leaders’ recent pledges of allegiance to the RSF could significantly intensify fighting and civilian targeting.

**Key escalation factors for 2025:**
– Continued siege warfare with multiple major battles documented
– Expansion of tribal involvement increasing ethnic targeting
– Sustained humanitarian crisis with over 11.7 million displaced[4]
– Regular military operations with aircraft, artillery, and large-scale ground combat

Grok response:

1. **Title:** ACLED Conflict Watchlist 2025: Sudan
**Description:** ACLED’s annual watchlist report tracking state-based violence, fatalities, and conflict trends in Sudan.
**Date:** 2025-09-15 (latest update as accessed on 2025-10-03).
**Summary:** This report estimates over 12,500 state-based conflict deaths in Sudan for 2025 as of mid-September, driven primarily by clashes between the Sudanese Armed Forces and Rapid Support Forces. It highlights a surge in violence during the third quarter, with data showing monthly averages of 1,500 deaths in August and September due to urban battles and ethnic targeting. Experts warn that without intervention, deaths could exceed prior years’ totals, emphasizing the need for international monitoring.
2. **Title:** War in Sudan: Humanitarian, Fighting, and Control Developments, September 2025
**Description:** Al Jazeera’s overview of the ongoing civil war, including ACLED-referenced casualty figures.
**Date:** 2025-09-30 (published 3 days ago).
**Summary:** The article reports ACLED data indicating at least 2,000 state-based conflict deaths in September 2025 alone, amid intensified fighting in Darfur and Khartoum. It notes that these deaths are predominantly from government-affiliated forces, with displacement exacerbating the crisis. Analysts predict continued escalation without ceasefires, based on ACLED’s real-time tracking.
3. **Title:** World Leaders Step Up Efforts Behind the Scenes at the U.N. to End the War in Sudan
**Description:** NBC News article on international responses to the Sudan conflict, citing ACLED for death toll estimates.
**Date:** 2025-09-26 (published about 1 week ago).
**Summary:** Drawing from ACLED, the piece states that at least 40,000 total conflict-related deaths have occurred in Sudan since the war began, with over 12,000 attributed to state-based violence in 2025. It discusses U.N. efforts to address the crisis, including calls for data-driven interventions. Experts from the U.N. emphasize the need for accurate tracking to prevent further fatalities.

I did not produce any forecasts, as my role is to provide context only.
[‘https://acleddata.com/africa/horn-of-africa/sudan/‘, ‘https://acleddata.com/conflict-watchlist-2025/sudan/‘]
Extracted resolution criteria links:

Claude Vision analysis response:
Looking at this ACLED data visualization for Sudan, I can extract the following relevant information about state-based conflict deaths:

## Current Data (as of early October 2025):
– **2025 Year-to-Date Fatalities**: 12,257 deaths reported
– This represents data through approximately 9 months of 2025 (January through early October)

## Historical Context:
– **2024 Full Year**: 26,027 fatalities
– **2023 Full Year**: 15,553 fatalities

GPT Multiple Choice JSON response:
ACLED is the resolution authority; their year-to-date (YTD) reporting is the primary source. ACLED’s public materials and the screenshots cited in my research indicate ~12.2k state-based conflict fatalities in Sudan through early October 2025 (12,257 YTD in the cited visualization). That means, as of today (2025-10-03), the 2025 YTD total is already above the 12,000 threshold. Unless there is a major retrospective reclassification or large downward data correction by ACLED, the final 2025 total reported by ACLED will therefore almost certainly be in the “More than 12000” bucket.

To land in any bin below 12,000 would require (a) the published YTD figure to be incorrect by many thousands, or (b) ACLED to retroactively reclassify or remove a very large number of events/fatalities — both unlikely given ACLED’s conservative, accumulating approach to event tallies.

Final probabilities:
Option_A: 1%
Option_B: 1%
Option_C: 1%
Option_D: 2%
Option_E: 8%
Option_F: 87%

Gemini Multiple Choice response:

According to the Vision Analysis of ACLED’s data visualization, the number of fatalities for 2025 year-to-date, as of early October 2025, already stands at 12,257 deaths. This figure is corroborated by the Social Media Research, which cites ACLED’s watchlist updates estimating “over 12,500 state-based conflict deaths in Sudan for 2025 as of mid-September”.

Historical context further reinforces this prediction. ACLED data shows 26,027 fatalities in 2024 and 15,553 in 2023, indicating a sustained and high-intensity conflict. The average monthly death rate for 2025, based on the 12,257 deaths over approximately nine months, is around 1,362 fatalities per month. If this rate continues for the remaining three months of 2025 (October, November, December), an additional ~4,086 deaths would be projected, bringing the total to approximately 16,343 deaths.

Given that the current reported figure for 2025 has already surpassed 12,000, all other categories are practically impossible.

Less than 1,000: 0%
Between 1000 and 3,000: 0%
Between 3000 and 5,000: 0%
Between 5000 and 8,000: 0%
Between 8000 and 12,000: 1%
More than 12,000: 99%

Anthropic response:
The most authoritative source is the ACLED data visualization showing 12,257 deaths through early October 2025 (approximately 9 months of data). This is already above the 12,000 threshold with roughly 3 months remaining in the year.

Recent reporting from September 2025 indicated approximately 2,000 deaths in that month alone, suggesting continued high intensity. The siege of El Fasher remains active with documented major engagements. The expansion of tribal involvement and RSF consolidation in certain regions could maintain or escalate violence levels.

Given we already have 12,257 confirmed deaths with 3 months remaining, falling below 12,000 is essentially impossible barring massive data revisions. The “Between 8000 and 12000” category is also extremely unlikely as it would require negative deaths or major downward revisions.

Less than 1,000: 0%
Between 1,000 and 3,000: 0%
Between 3,000 and 5,000: 0%
Between 5,000 and 8,000: 1%
Between 8,000 and 12,000: 2%
More than 12,000: 97%

GPT Multiple Choice JSON response:
Rationale and key evidence
The authoritative resolution source is ACLED’s Sudan pages. The available ACLED-derived figures shown in the research material and vision-analysis for early October 2025 report roughly 12,257 state-based conflict deaths in Sudan for 2025 year-to-date.

Uncertainty and error sources: my main uncertainty is not the direction of the trend (which is high and sustained) but the possibility of data/mapping errors or later ACLED reclassification.

How I translated this into probabilities
Given (a) an ACLED YTD tally already above 12k as of early October 2025, (b) the remainder of the year still containing roughly three months when nontrivial fatalities are likely, and (c) the low historical likelihood of large downward revisions that would drop a present >12k count below the 12k threshold, the vast majority of probability mass should go to “More than 12000.” I nevertheless allocate small probabilities across other bins to reflect the nonzero chance of misreading, data revision, or highly unusual rapid de-escalation combined with ACLED reclassification. I avoid absolute-certainty extremes because ACLED data and counting definitions can be updated and because the research material could be misinterpreted.

Final probability distribution
Less than 1,000: 1%
Between 1,000 and 3,000: 1%
Between 3,000 and 5,000: 1%
Between 5,000 and 8,000: 2%
Between 8,000 and 12,000: 5%
More than 12,000: 90%

Model’s Answer:
The most authoritative source is the ACLED data visualization showing 12,257 deaths through early October 2025 (approximately 9 months of data). This is already above the 12,000 threshold with roughly 3 months remaining in the year. The current 2025 trajectory would result in roughly 16,000-17,000 deaths, representing a ~35-40% decrease from 2024 levels.

Recent reporting from September 2025 indicated approximately 2,000 deaths in that month alone, suggesting continued high intensity. The siege of El Fasher remains active with documented major engagements. The expansion of tribal involvement and RSF consolidation in certain regions could maintain or escalate violence levels.

Three critical assumptions that could alter this forecast:
1. **Assumption of continued monthly average (70% confidence in validity)**: I’m assuming the ~1,362 deaths/month average will roughly continue. However, this could fail if there’s either a major escalation (new offensive campaigns, breakdown of humanitarian corridors) or de-escalation (ceasefire agreements, international intervention). Historical Q4 patterns show high variability.
2. **Assumption of accurate current data (85% confidence)**: The 12,257 figure from ACLED appears current as of early October, but there could be reporting delays or revisions. ACLED sometimes updates historical figures as more information becomes available from conflict zones. A 10-20% revision wouldn’t be unprecedented.
3. **Assumption of no major conflict transformation (75% confidence)**: I’m assuming the conflict continues in its current form. However, potential game-changers include: collapse of either SAF or RSF positions leading to rapid territorial changes, external military intervention, or successful peace negotiations. The UN efforts mentioned in September could materialize into concrete action.

Given we already have 12,257 confirmed deaths with 3 months remaining, falling below 12,000 is essentially impossible barring massive data revisions.

Less than 1,000: 0%
Between 1,000 and 3,000: 0%
Between 3,000 and 5,000: 0%
Between 5,000 and 8,000: 1%
Between 8,000 and 12,000: 2%
More than 12,000: 97%

Past forecasts by Phil’s and Jeremy ’s bots —>

Below, a forecast of state-based conflict fatalities for Sudan in the VIEWS machine forecasting competition, which is an aggregate of the forecasts of twenty competitors. Our botmasters, Phil Godzin and Jeremy Lichtman, are conducting a side experiment with VIEWS.