How many state-based conflict deaths will be reported by ACLED in 2025?
Full details from what Phillip Godzin’s pgodzinai predicts today:
* Less than 100: 2%
* Between 100 and 250: 13%
* Between 250 and 500: 32%
* Between 500 and 1000: 35%
* Greater than 1000: 18%
Perplexity response:
{think}
Okay, let’s tackle this question step by step. The user wants to know how many state-based conflict deaths ACLED will report in Sudan for 2025, with specific options provided. First, I need to check the search results for any ACLED data related to Sudan in 2025.
Looking at the search results, the most recent mention of Sudan is in result [1], dated May 11, 2025. It states that more than 30,000 people have been killed in Sudan, with the real number much higher. This is a significant number, but I need to confirm if this is from ACLED. However, the quote doesn’t explicitly mention ACLED, so I can’t cite that as ACLED data.
Next, results [2], [4], and [5] mention ACLED data. For example, [2] talks about an estimated 6,000 civilian deaths between January and August 2024, but that’s for the African Union’s roadmap. [4] refers to ACLED data on conflict victims in Burkina Faso, mentioning over half of the deaths occurred in the past three years. [5] discusses Burkina Faso’s situation with ACLED data showing deaths before and after a coup, but this isn’t relevant to Sudan.
The key point here is that none of the provided search results specifically reference ACLED’s 2025 data for Sudan. The only Sudan-related data is from [1], which doesn’t attribute the numbers to ACLED. Without direct ACLED sources or more recent data from 2025, it’s challenging to pinpoint the exact figure.
The user also provided links to ACLED’s Sudan pages, but since those aren’t included in the search results, I can’t access them. The available data from the search results doesn’t give a reliable figure for 2025 Sudan-specific state-based conflict deaths from ACLED. Therefore, based on the given information, there’s insufficient data to determine the exact number. However, result [1] mentions over 30,000 deaths in Sudan as of May 2025, which would fall into the “More than 12000” category. But this isn’t explicitly linked to ACLED, so there’s uncertainty.
Given the ambiguity, the answer should indicate that there’s no direct ACLED data in the provided sources but note the high death toll mentioned, advising caution due to potential discrepancies between sources.
Below is a synthesized analysis of relevant data points and contextual factors regarding state-based conflict deaths in Sudan for 2025, based on available sources.
—
### **Key Data and Context**
1. **Sudan’s Ongoing Crisis**
– As of May 11, 2025, Sudan is described as experiencing one of the world’s worst humanitarian crises, with over 30,000 people reported killed. However, the “real number is much higher”[1].
– This aligns with broader regional trends, as ACLED data cited elsewhere shows heightened conflict-related fatalities in Africa, though Sudan-specific 2025 figures are not explicitly broken out[2][5].
2. **ACLED Data Limitations**
– While ACLED tracks conflict deaths globally, the provided search results do not include explicit Sudan-specific ACLED data for 2025. The **2025 Human Rights Roadmap for the African Union**[2] cites 6,000 civilian deaths in conflict-related violence between January–August 2024, but this refers to a regional aggregate, not Sudan alone.
– Burkina Faso’s conflict, tracked by ACLED, saw a doubling of deaths post-coup[5], but this is unrelated to Sudan.
3. **Resolution Criteria Challenges**
– The resolution criteria rely on ACLED’s Sudan-specific figures for 2025, which are not directly accessible via the provided links. The most recent Sudan-specific data point (30,000+ deaths)[1] does not specify whether these fatalities are categorized as “state-based” or include broader conflict-related violence.
– ACLED defines “state-based conflict” narrowly, which may exclude deaths from non-state actors or indirect violence, potentially lowering reported totals compared to aggregations like[1].
4. **Expert Opinions and Propaganda Context**
– Analysts note heightened instability in neighboring countries, such as Burkina Faso, where over 60% of the country is outside government control[5]. While not directly about Sudan, this reflects broader regional security challenges.
– In Sudan, the conflict’s severity and underreporting are acknowledged[1], suggesting ACLED’s figures may underestimate actual deaths.
—
### **Options Analysis**
– **”Less than 1000″** or **”Between 1000–3000″**: Unlikely, given the scale of violence described[1].
– **”Between 3000–5000″**, **”5000–8000″**, **”8000–12,000″**: Possible but contingent on ACLED’s methodology excluding non-state actor deaths.
– **”More than 12,000″**: Aligns with the >30,000 total deaths cited[1], assuming “state-based” accounts for a significant portion.
—
### **Critical Uncertainties**
1. **Methodological Differences**: ACLED’s state-based definition may exclude deaths from non-government actors, potentially lowering totals compared to broader conflict mortality counts[1][5].
2. **Data Gaps**: No 2025-specific ACLED figures for Sudan are available in the provided sources. Historical data (e.g., pre-2025) is absent in the search results, complicating trend analysis.
—
### **Conclusion**
Without direct access to ACLED’s 2025 Sudan-specific data, the answer hinges on interpreting the highest available figure (>30,000 total deaths)[1] and the criteria’s focus on “state-based” fatalities. This suggests a likely resolution in **”More than 12,000″**, though caution is advised due to potential underreporting and methodological exclusions.
AskNews Deep Usage:
prompt_tokens=19566 completion_tokens=2539 total_tokens=0
AskNews Deep Search Response:
To forecast the number of state-based conflict deaths in Sudan in 2025 reported by ACLED, recent reports indicate severe escalation but lack explicit ACLED data for 2025. Here’s the synthesized context:
1. **Escalating Violence**: Recent attacks in Darfur and Khartoum show significant fatalities. In May 2025 alone, **542 civilian deaths were reported in North Darfur over three weeks** [25][30][34][35], suggesting a monthly rate of ~700 deaths if sustained. Extrapolated annually, this implies ~8,400 deaths, though localized to one region.
2. **Conflict Timeframe**: The war began in April 2023. Earlier sources cite **cumulative death tolls ranging from 20,000 (UN) to 150,000 (scholars) by mid-2024** [4][19]. If violence persists at similar intensity, 2025 deaths could align with **“Between 5,000 and 8,000”** or higher.
3. **Discrepancies in Reporting**:
– UN/local authorities: **Over 20,000 killed** by May 2025 (cumulative since 2023) [4][17][19].
– Academic estimates: Up to **130,000–150,000 deaths** by 2025 [4][18][19].
– Recent localized surges: At least **40+ daily deaths** in El Fasher [24][30][34].
4. **ACLED Context**: While ACLED’s 2025 Sudan data isn’t directly cited in the search results, its 2024 global estimate of **223,000 conflict deaths** [21] underscores Sudan’s disproportionate share due to the ongoing crisis.
**Key Limitations**:
– Most sources aggregate deaths from 2023–2025.
– Recent localized reports (e.g., Darfur) imply high lethality but may not reflect nationwide totals.
– Displacement and famine (24M facing hunger [3]) complicate fatality tracking.
**Conclusion**: If current violence persists, ACLED’s 2025 death toll may fall in **“Between 5,000 and 8,000”** based on localized rates, but the upper brackets (“Between 8,000 and 12,000” or “More than 12,000”) are plausible given historical underreporting and escalating attacks [1][4][19][24][30].
GPT News Summary:
To help assess the likely number of state-based conflict deaths in Sudan in 2025 reported by ACLED, recent reports provide critical insights into the ongoing conflict dynamics and humanitarian situation.
1. **Conflict and Humanitarian Impact**: As of May 2025, the Sudanese conflict remains severe. The number of internally displaced persons due to the conflict is reported at 12 million, and nearly 30,000 people have been killed (Alleyoop, May 14, 2025). Additionally, since the conflict intensified in 2023, there have been over 120 deaths due to bombings in just a few days in January 2025 (Agensir, January 17, 2025).
2. **Recent Violence and Civilians at Risk**: The escalation continues with over 40,000 refugees fleeing to Chad in one month due to violence in Darfur (AA, May 15, 2025; UN, May 14, 2025). Reports indicate a precarious security situation, especially in Darfur, with ongoing fighting causing heavy civilian casualties (UN, May 14, 2025).
3. **Human Rights Abuses**: Systematic attacks on women and children are evident, with at least 330 cases of conflict-related sexual violence documented in 2025, although the real number could be much higher due to underreporting (ReliefWeb, May 15, 2025).
4. **Conflict Intensity and Fatalities**: As of April 2025, the conflict had resulted in at least 32,973 deaths, with half of the fatalities occurring in Khartoum and North Darfur (Al Jazeera, April 14, 2025). This underscores the severe and ongoing impact of the conflict on the civilian population.
5. **Cholera and Public Health Crisis**: A choleral outbreak has been reported, with 2,307 deaths in 2025, mainly affecting Sudan, South Sudan, Angola, and the Democratic Republic of the Congo. This public health crisis exacerbates the overall humanitarian emergency (La Razón, May 15, 2025).
6. **International Aid and Food Security**: Sudan faces a severe food insecurity crisis, exacerbated by a decline in international aid. Over 150,000 deaths related to hunger are reported, with 25 million people facing hunger due to the ongoing civil war (One Green Planet, May 16, 2025). The UN warns about worsening food security due to reduced humanitarian funding (El Nacional, May 16, 2025).
Overall, the ongoing conflict, compounded by humanitarian crises such as food insecurity and public health threats like cholera, suggests a high likelihood of continued high numbers of conflict-related deaths throughout the rest of 2025. These factors must be carefully considered in forecasting Sudan’s conflict deaths for the remainder of the year.
Extracted resolution criteria links:
[]
GPT Multiple Choice JSON response:
Option_A: 1
Option_B: 4
Option_C: 10
Option_D: 22
Option_E: 36
Option_F: 27
probability_yes_per_category: {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.1, ‘Between 5000 and 8000’: 0.22, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.27}
________________________
Anthropic response:
Given this analysis, here’s my probability distribution for ACLED-reported state-based conflict deaths in Sudan for 2025:
Option_A: 5
Option_B: 10
Option_C: 15
Option_D: 30
Option_E: 25
Option_F: 15
probability_yes_per_category: {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.15}
________________________
Anthropic response:
Given this analysis, and considering:
– We need to forecast May-December 2025 (7.5 months)
– Recent SAF territorial gains could reduce battle deaths but increase civilian targeting
– Historical monthly averages suggest approximately 1,000-1,500 deaths per month
– Significant underreporting is likely
Probability Distribution:
Option_A: 5
Option_B: 10
Option_C: 15
Option_D: 35
Option_E: 25
Option_F: 10
probability_yes_per_category: {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.35, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.1}
GPT Multiple Choice JSON response:
Less than 1000: 1
Between 1000 and 3000: 3
Between 3000 and 5000: 6
Between 5000 and 8000: 18
Between 8000 and 12000: 35
More than 12000: 37
probability_yes_per_category: {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.03, ‘Between 3000 and 5000’: 0.06, ‘Between 5000 and 8000’: 0.18, ‘Between 8000 and 12000’: 0.35, ‘More than 12000’: 0.37}
________________________
Anthropic response:
.
Forecast Analysis:
– The recent capture of Khartoum by SAF could lead to a temporary reduction in urban casualties but might shift violence to other regions
– ACLED’s historical monthly fatality rates suggest a baseline of several thousand deaths per quarter
– The intensification of fighting in early 2025 indicates the conflict remains highly active
– Given ACLED’s conservative methodology, their reported numbers are likely to be significantly lower than actual deaths
For the remaining ~7.5 months of 2025 (mid-May through December), considering current intensity levels and historical patterns, I estimate:
Option_A: 5
Option_B: 15
Option_C: 25
Option_D: 30
Option_E: 15
Option_F: 10
probability_yes_per_category: {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.15, ‘Between 3000 and 5000’: 0.25, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.15, ‘More than 12000’: 0.1}
GPT Multiple Choice JSON response:
Less than 1000: 2
Between 1000 and 3000: 6
Between 3000 and 5000: 11
Between 5000 and 8000: 20
Between 8000 and 12000: 29
More than 12000: 32
probability_yes_per_category: {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.11, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.29, ‘More than 12000’: 0.32}
GPT Multiple Choice JSON response:
Less than 1000: 2
Between 1000 and 3000: 6
Between 3000 and 5000: 12
Between 5000 and 8000: 20
Between 8000 and 12000: 30
More than 12000: 30
probability_yes_per_category: {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.12, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.3, ‘More than 12000’: 0.3}
GPT Multiple Choice JSON response:
Sudan’s civil war between the SAF and RSF continues to generate several-hundred to more-than-one-thousand conflict fatalities most weeks. ACLED’s methodology is conservative but still registered roughly 12–15 k deaths from Apr-2023 → Apr-2024; 2024 almost certainly ended in the low- to mid-teens. Through mid-May 2025, a plausible cumulative 4–6 k deaths have already been coded for 2025. For the year to finish below 8 k, violence would have to fall to <350 deaths per month for the rest of 2025—historically unprecedented since the war began and unlikely while major battles continue around El-Fasher, Khartoum and Kordofan. A mid-range outcome (8–12 k) would require an average of roughly 900 deaths per month from mid-May onward—plausible if intensity slowly ebbs or reporting gaps widen. Continuation of the current tempo or any renewed escalation (which recent trends in Darfur support) would push the 2025 total over 12 k. Because ACLED’s state-based category excludes some RSF-on-civilian atrocities, even high real-world fatality counts do not guarantee extremely high ACLED tallies, so a non-trivial chance remains for the 8–12 k bin.
Balancing these considerations, with residual uncertainty about ACLED coding practice and possible cease-fire scenarios, yields the following probability distribution:
Less than 1000: 1
Between 1000 and 3000: 4
Between 3000 and 5000: 7
Between 5000 and 8000: 13
Between 8000 and 12000: 36
More than 12000: 39
probability_yes_per_category: {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.07, ‘Between 5000 and 8000’: 0.13, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.39}
Gemini Multiple Choice response:
The conflict in Sudan, which escalated significantly in April 2023, has continued with high intensity into 2025. The question requires forecasting the number of *state-based conflict deaths* reported by ACLED for the calendar year 2025. As of mid-May 2025, approximately 4.5 months of the year have passed.
Recent reports indicate severe and ongoing violence. For instance, one research assistant highlighted that in May 2025 alone, over 542 civilian deaths were reported in North Darfur over just three weeks. If such events are classified by ACLED as state-based conflict deaths and are indicative of the violence in other hotspots (like Khartoum, which historically accounts for a large share of fatalities), the monthly death toll could be substantial. Another report mentioned over 120 deaths from bombings in a few days in January 2025.
Estimating the deaths from January 1 to May 15, 2025, is crucial. If we conservatively assume that ACLED records, for example, an average of 1,000 to 2,000 state-based conflict deaths per month across Sudan based on the reported intensity, then 4.5 months would yield approximately 4,500 to 9,000 deaths already in 2025. This initial estimate already pushes the likely annual total towards the higher brackets.
Projecting for the remaining 7.5 months of 2025, if the conflict continues at a similar intensity (e.g., 1,000-2,000 deaths per month), an additional 7,500 to 15,000 deaths could occur.
* If Jan-May saw
Task failed with error: Number of options (6) does not match number of probabilities (4)
Traceback (most recent call last):
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 107, in fetch_gemini_multiple_choice_prediction
return await fetch_sync(get_gemini_multiple_choice_prediction, title, perplexity_summary_report, asknews_summary, asknews_deep_research, vision_analysis, today, resolution_date, background, fine_print, resolution_criteria, options)
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 61, in fetch_sync
return await asyncio.get_event_loop().run_in_executor(
File “/opt/hostedtoolcache/Python/3.10.17/x64/lib/python3.10/concurrent/futures/thread.py”, line 58, in run
result = self.fn(*self.args, **self.kwargs)
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/main.py”, line 57, in run_sync
return future.result(timeout=180)
File “/opt/hostedtoolcache/Python/3.10.17/x64/lib/python3.10/concurrent/futures/_base.py”, line 458, in result
return self.__get_result()
File “/opt/hostedtoolcache/Python/3.10.17/x64/lib/python3.10/concurrent/futures/_base.py”, line 403, in __get_result
raise self._exception
File “/opt/hostedtoolcache/Python/3.10.17/x64/lib/python3.10/concurrent/futures/thread.py”, line 58, in run
result = self.fn(*self.args, **self.kwargs)
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/gemini_service_2.py”, line 100, in get_gemini_multiple_choice_prediction
probability_yes_per_category = generate_multiple_choice_forecast(options, option_probabilities)
File “/home/runner/work/metaculus-forecasting-llm/metaculus-forecasting-llm/src/utils.py”, line 56, in generate_multiple_choice_forecast
raise ValueError(
ValueError: Number of options (6) does not match number of probabilities (4)
All predictions:
[{‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.15}, {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.15, ‘Between 3000 and 5000’: 0.25, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.15, ‘More than 12000’: 0.1}, {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.12, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.3, ‘More than 12000’: 0.3}, {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.11, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.29, ‘More than 12000’: 0.32}, {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.35, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.1}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.1, ‘Between 5000 and 8000’: 0.22, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.27}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.03, ‘Between 3000 and 5000’: 0.06, ‘Between 5000 and 8000’: 0.18, ‘Between 8000 and 12000’: 0.35, ‘More than 12000’: 0.37}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.07, ‘Between 5000 and 8000’: 0.13, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.39}]
Filtered predictions:
[{‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.12, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.3, ‘More than 12000’: 0.3}, {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.11, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.29, ‘More than 12000’: 0.32}, {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.35, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.1}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.1, ‘Between 5000 and 8000’: 0.22, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.27}]
Averaged predictions:
{‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.12, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.3, ‘More than 12000’: 0.3}
Median reasoning:
All Predictions:
[{‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.15}, {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.15, ‘Between 3000 and 5000’: 0.25, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.15, ‘More than 12000’: 0.1}, {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.12, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.3, ‘More than 12000’: 0.3}, {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.11, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.29, ‘More than 12000’: 0.32}, {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.35, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.1}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.1, ‘Between 5000 and 8000’: 0.22, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.27}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.03, ‘Between 3000 and 5000’: 0.06, ‘Between 5000 and 8000’: 0.18, ‘Between 8000 and 12000’: 0.35, ‘More than 12000’: 0.37}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.07, ‘Between 5000 and 8000’: 0.13, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.39}]
Median Reasoning:
EXTRACTED_PROBABILITIES: [5, 10, 15, 35, 25, 10]
Model’s Answer:
Given this analysis, and considering:
– We need to forecast May-December 2025 (7.5 months)
– Recent SAF territorial gains could reduce battle deaths but increase civilian targeting
– Historical monthly averages suggest approximately 1,000-1,500 deaths per month
– Significant underreporting is likely
Probability Distribution:
Option_A: 5
Option_B: 10
Option_C: 15
Option_D: 35
Option_E: 25
Option_F: 10
=== Forecast Result ===
{
“probability_yes”: null,
“probability_yes_per_category”: {
“Less than 1000”: 0.024999999999999998,
“Between 1000 and 3000”: 0.065,
“Between 3000 and 5000”: 0.12,
“Between 5000 and 8000”: 0.2425,
“Between 8000 and 12000”: 0.3,
“More than 12000”: 0.2475
},
“continuous_cdf”: null
}
=== Reasoning ===
All Predictions:
[{‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.15}, {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.15, ‘Between 3000 and 5000’: 0.25, ‘Between 5000 and 8000’: 0.3, ‘Between 8000 and 12000’: 0.15, ‘More than 12000’: 0.1}, {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.12, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.3, ‘More than 12000’: 0.3}, {‘Less than 1000’: 0.02, ‘Between 1000 and 3000’: 0.06, ‘Between 3000 and 5000’: 0.11, ‘Between 5000 and 8000’: 0.2, ‘Between 8000 and 12000’: 0.29, ‘More than 12000’: 0.32}, {‘Less than 1000’: 0.05, ‘Between 1000 and 3000’: 0.1, ‘Between 3000 and 5000’: 0.15, ‘Between 5000 and 8000’: 0.35, ‘Between 8000 and 12000’: 0.25, ‘More than 12000’: 0.1}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.1, ‘Between 5000 and 8000’: 0.22, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.27}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.03, ‘Between 3000 and 5000’: 0.06, ‘Between 5000 and 8000’: 0.18, ‘Between 8000 and 12000’: 0.35, ‘More than 12000’: 0.37}, {‘Less than 1000’: 0.01, ‘Between 1000 and 3000’: 0.04, ‘Between 3000 and 5000’: 0.07, ‘Between 5000 and 8000’: 0.13, ‘Between 8000 and 12000’: 0.36, ‘More than 12000’: 0.39}]
Median Reasoning:
EXTRACTED_PROBABILITIES: [5, 10, 15, 35, 25, 10]
Model’s Answer:
Given this analysis, and considering:
– We need to forecast May-December 2025 (7.5 months)
– Recent SAF territorial gains could reduce battle deaths but increase civilian targeting
– Historical monthly averages suggest approximately 1,000-1,500 deaths per month
– Significant underreporting is likely
Probability Distribution:
Option_A: 5
Option_B: 10
Option_C: 15
Option_D: 35
Option_E: 25
Option_F: 10
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.
Below, control of terrain in Sudan. Source: https://www.understandingwar.org/backgrounder/africa-file-april-10-2025-rsf%E2%80%99s-northern-sudan-campaign
